A cross-platform flower order fulfillment status alignment and exception reassignment method

Abstract

The application relates to the technical field of cross-platform flower order, in particular to a cross-platform flower order performance state alignment and abnormal reassignment method.The method comprises the following steps: obtaining a user platform order information set and a merchant platform commodity information set; based on the user platform order information set, the merchant platform commodity information set is combined to obtain a user-merchant performance deviation information set; a delivery platform state information set is obtained; according to the delivery platform state information set, the user-merchant performance deviation information set is combined to obtain a delivery risk assessment information set; the user-merchant performance deviation information set and the delivery risk assessment information set are integrated to perform abnormal reassignment level division, and a cross-platform flower order abnormal reassignment strategy is output and executed.The application can realize state alignment of the whole flower performance process, quantize quality risks and execute differentiated reassignment strategies, and improve the processing efficiency of abnormal orders and user experience.

Description

Technical Field

[0001] This application relates to the field of cross-platform flower order technology, and in particular to a method for aligning the fulfillment status and resending abnormal flower orders across platforms. Background Technology

[0002] In the existing flower supply chain system, the order fulfillment process usually involves multiple stages, including user ordering, merchant processing and warehousing, and logistics and distribution. In order to ensure that the flowers can be delivered to users in the best condition, the industry generally adopts a multi-platform data collaboration approach, that is, the user platform records order requirements, the merchant platform manages the processing technology and storage environment of the flowers, and the distribution platform monitors the transportation route and real-time status.

[0003] However, in the actual fulfillment of flower orders, as a special commodity with biological activity and susceptible to external environmental influences, the final quality of flowers often deviates unavoidably from user expectations. This is especially true when dealing with flowers that have undergone special artificial processing. Due to the interplay of the complexity of processing techniques, the uncontrollability of natural aging processes, and the variability of transportation environments, key sensory indicators such as color, shape, and odor of the flowers are prone to fluctuations. Currently, the industry lacks a universal mechanism that can quantify and assess such quality deterioration caused by multiple factors, and match differentiated handling strategies across the entire supply chain. This makes it difficult to achieve efficient and reasonable responsibility determination and service compensation when non-standardized quality anomalies occur in some orders, affecting the overall service experience and operational efficiency. Summary of the Invention

[0004] This application provides a method for aligning the fulfillment status and resending abnormal flower orders across platforms to solve the above-mentioned problems. The method includes: Obtain the user platform order information set and the merchant platform product information set. Based on the user platform order information set and the merchant platform product information set, analyze the perceptual deviation characteristics of floral materials caused by human processing and natural evolution, as well as the changes of floral materials under physical combination and structural constraints, to obtain the user-merchant fulfillment deviation information set. Obtain the delivery platform status information set, and based on the delivery platform status information set and the user-merchant fulfillment deviation information set, analyze the retention capacity of the flowers during the delivery process and the characteristics of quality re-change caused by environmental factors to obtain the delivery risk assessment information set. Integrate the user-merchant fulfillment deviation information set and the delivery risk assessment information set, classify abnormal reassignment levels, and output and execute cross-platform flower order abnormal reassignment strategies.

[0005] Optionally, the process of constructing the user-merchant performance deviation information set includes: The user platform order information set includes the type of bouquet and the quantity of flowers. The merchant platform's product information set includes information on flower material processing and flower material warehousing. Based on the type of bouquet, the quantity of flowers, and the processing information of the flowers, the effects of packaging methods on the size and shape of the flowers, as well as the effects of spraying on the color and smell of the flowers, are analyzed to obtain the human-induced size and shape deviations and the spraying color and smell deviations. Based on the quantity of flowers, the type of bouquet, and the flower storage information, the effects of storage temperature and humidity on the freshness and natural color of flowers are analyzed to obtain the natural evolution color deviation. By integrating the human-caused size and shape deviations, the color and odor deviations of the sprayed paint, and the color deviations of natural evolution, a user-merchant performance deviation information set is constructed.

[0006] Optionally, the process of constructing the artificial size and shape deviation includes: Based on the quantity of flowers and the processing information of the flowers, the physical compression characteristics of the external binding and wrapping methods on the original extended state of the flowers, as well as the physical restrictions of the internal space of the packaging container on the spreading angle of the flowers, are analyzed to obtain information on compression distortion. Based on the types of bouquets and the information on compression distortion, the differential tolerance characteristics of stem flexibility and flower fragility under packaging compression of different flowers are analyzed to obtain compression tolerance information. By integrating the pressure distortion information and the compression tolerance information, the artificial dimensional and morphological deviation is constructed.

[0007] Optionally, the process of constructing the color and odor deviation of the sprayed color includes: Based on the types of bouquets and the processing information of the flowers, the differences in the adhesion ability of the surface texture of the petals of different flowers to the sprayed pigments, as well as the masking characteristics of the original natural color of the flowers after the pigments are applied, are analyzed to obtain the color coverage deviation. Based on the quantity of the flowers and the processing information of the flowers, the uniform spraying characteristics of the sprayed pigments on a batch of flowers are analyzed, as well as the odor variation characteristics caused by the mixture of the residual odor after the solvent in the pigments evaporates and the fragrance of the flowers themselves, to obtain the odor mixing deviation. The color coverage deviation and the odor mixing deviation are integrated to construct the spray color odor deviation.

[0008] Optionally, the process of constructing the natural evolutionary color deviation includes: Based on the quantity of flowers and the storage information of flowers, the oxidation and discoloration characteristics of the flower cuts and epidermis due to contact with air are analyzed to obtain oxidation and discoloration information. Based on the oxidation discoloration information and the type of bouquet, the differences in the stability of petal pigments and the thickness of the epidermal wax layer of different floral materials are analyzed to determine the differential sensitivity characteristics of the color darkening and spot distribution caused by oxidation reaction, thus obtaining the natural evolution color deviation.

[0009] Optionally, based on the delivery platform status information set and the user-merchant fulfillment deviation information set, the ability of flowers to maintain their quality during delivery and the characteristics of quality re-change caused by environmental factors are analyzed to obtain a delivery risk assessment information set, including: The delivery platform status information set includes delivery route information, delivery time period, and delivery packaging type; Based on the delivery route information and the human-caused size and shape deviations, the further impact of transportation bumps on the size and shape of the flowers is analyzed to obtain transportation deformation information; Based on the transportation deformation information, combined with the delivery time period and the natural evolution color deviation, the influence of temperature changes during the delivery time period on the quality of the flowers is analyzed to obtain the quality change information during the time period; Based on the time period quality change information, combined with the delivery packaging type and the color and odor deviation of the sprayed paint, the influence of the packaging environment on the color and odor of the flowers is analyzed to obtain the delivery risk assessment information set.

[0010] Optionally, the process of constructing the transport deformation information includes: Based on the delivery route information, the vertical vibration frequency characteristics caused by differences in road surface smoothness in the transportation route, as well as the horizontal inertial characteristics during vehicle start-stop and turning processes, are analyzed to obtain bump and impact information. Based on the bump and impact information, combined with the human-induced size and shape deviation, the characteristics of flower head displacement caused by external impact on the basis of the original physical compression are analyzed to obtain the bouquet distortion information. Based on the bouquet distortion information, the mutual compression and friction characteristics between individual flower materials caused by displacement, as well as the curling and shedding trends of petal edges caused by collisions, are analyzed to obtain the transport deformation information.

[0011] Optionally, the process of constructing the time-period quality change information includes: Based on the transport deformation information, the damage to the epidermis of the compressed parts of the flower material and the water loss characteristics caused by the obstruction of the water supply channels inside the flower material are analyzed to obtain the flower material damage information. Based on the information on damaged flowers and the delivery time period, we analyzed the characteristics of changes in the respiration intensity of flowers caused by changes in ambient temperature during the delivery process, as well as the characteristics of accelerated water loss from damaged areas, to obtain physiological loss information. Based on the physiological depletion information and the natural evolutionary color deviation, the color deepening phenomenon of the damaged parts on the basis of the original oxidative discoloration and the color darkening phenomenon of the healthy parts due to water loss are analyzed to obtain the quality change information of the time period.

[0012] Optionally, based on the time-period quality change information, combined with the delivery packaging type and the color and odor deviation of the spray paint, the influence of the packaging environment on the color and odor of the flowers is analyzed to obtain the delivery risk assessment information set, including: Based on the time period quality change information and the delivery packaging type, the limiting characteristics of the light transmittance and air permeability of the packaging material on the maintenance of the flower color, as well as the shielding or mixing effect of the packaging material's own adsorption or release properties on the flower odor, are analyzed to obtain interactive influence information. Based on the interaction information and the color and odor deviation of the sprayed pigment, the secondary mixing and crosstalk characteristics of the residual solvent volatiles of the sprayed pigment and the natural fragrance of the flowers in the packaging space are analyzed to obtain the mixing crosstalk information. The interaction impact information and the mixed crosstalk information are integrated to construct the delivery risk assessment information set.

[0013] Optionally, the integration of the user-merchant fulfillment deviation information set and the delivery risk assessment information set, the classification of abnormal reassignment levels, and the output and execution of cross-platform flower order abnormal reassignment strategies include: Based on the user-merchant fulfillment deviation information set and the delivery risk assessment information set, the expected quality degradation limit of the flowers when they arrive at the user is analyzed under the superposition of the transportation deformation information, the time period quality change information and the interaction impact information, so as to obtain comprehensive fulfillment risk information. Based on the comprehensive performance risk information, the corresponding abnormal reassignment level is determined, and the cross-platform flower order abnormal reassignment strategy is output and executed. The abnormal reassignment levels include Level 1 reassignment for flowers that have sensory differences but still meet basic delivery standards, Level 2 reassignment for flowers whose physical structural integrity is damaged, and Level 3 reassignment for flowers that have lost their ornamental value and have an odor. When the order is determined to be a Level 1 reassignment, a reassurance strategy is implemented by sending the user an explanation of the difference between the actual flowers and the order, along with a coupon. When the reassignment is determined to be a Level 2 reassignment, a compensation strategy of providing the user with a partial refund and re-delivering replacement flowers will be implemented; If the reassignment is determined to be at level three, a full refund and cancellation of the original order will be implemented.

[0014] Through the above solution, this application achieves the following beneficial effects: A comprehensive quality assessment system was established, encompassing the entire supply chain from merchant processing to last-mile delivery. Correlation analysis between user platform order information sets and merchant platform product information sets quantified the initial perceptual deviations of flowers under the influence of human processing and natural evolution. The introduction of delivery platform status information sets further dynamically assessed the impact of the logistics environment on the preservation capabilities of flowers and their subsequent quality changes. Ultimately, by integrating user-merchant fulfillment deviation information sets with delivery risk assessment information sets, the final delivery quality of flowers can be objectively predicted. Based on this, differentiated re-delivery strategies for abnormal deliveries can be implemented, effectively resolving the re-delivery chaos caused by unclear responsibility definitions in existing technologies. This achieves intelligent alignment of flower order fulfillment status and proactive risk management.

[0015] It provides data support for subsequent delivery risk assessment and abnormal re-delivery decisions, and can identify the specific responsible links and risk sources that lead to abnormal performance based on multi-dimensional deviation quantification results.

[0016] It can reflect the systematic impact of human processing and packaging on the final presentation of floral materials, thereby providing a quantitative basis for cross-platform contract fulfillment alignment and subsequent abnormal reassignment decisions, ensuring that decisions can take into account both the preservation of morphological aesthetics and the integrity of plant physiology.

[0017] It enables quantitative monitoring of the quality of sprayed flowers, thereby effectively reducing the risk of customer complaints caused by color difference or odor.

[0018] By introducing species difference analysis, we avoid evaluating all flowers as homogeneous materials and can effectively distinguish between natural evolution caused by the biological characteristics of the flowers and abnormal deterioration caused by improper storage.

[0019] By combining user-merchant fulfillment deviation information sets and using the initial defects accumulated in front-end processing and warehousing as initial state values, the dynamic evolution process of floral materials in a specific packaging micro-environment can be simulated. This end-to-end coupled analysis mechanism enables the final output delivery risk assessment information set to comprehensively reflect the quality maintenance capability of floral materials in the last-mile delivery stage, thereby providing data support for subsequent abnormal re-delivery strategies.

[0020] By coupling macroscopic road condition incentives with microscopic physical damage to floral materials, the bumps and impacts generated by road surface smoothness and driving behavior are applied to floral materials with pre-existing human-induced shape deviations, resulting in flower head displacement and overall distortion. This leads to internal compression friction and edge collisions, ultimately forming transport deformation information that can be quantified and assessed. This collaborative process makes the previously unpredictable physical losses transparent and controllable, providing a basis for aligning the fulfillment status of cross-platform flower orders.

[0021] By organically combining the implicit water transport barriers caused by physical compression, the accelerated metabolic consumption driven by temperature, and the visual manifestation of oxidation reactions, the dynamic quality degradation trajectory of floral materials in the critical stage of delivery is depicted. This analysis method based on multi-source information fusion can effectively capture physiological quality deterioration that is difficult to detect by single physical state monitoring, ensuring the comprehensiveness and timeliness of delivery risk assessment, thereby providing data support for formulating abnormal relocation strategies.

[0022] This has enabled a shift from a single physical damage assessment to a comprehensive assessment encompassing physical, chemical, and sensory dimensions. In particular, it addresses the unique chemical stability issues of sprayed floral materials, providing data support for the subsequent development of abnormal relocation strategies.

[0023] By deeply coupling user-merchant fulfillment deviation information set with delivery risk assessment information set, the system enables quantitative assessment and tiered handling of flower order fulfillment status. The calculation of expected quality degradation limits on the technology side provides an objective and quantifiable basis for decision-making on the business side's abnormal resending strategy. This changes the passive situation of relying on subjective judgment by human customer service or handling only after user complaints. This data-driven tiered resending mechanism can not only quickly respond to different levels of fulfillment anomalies, reducing the platform's customer service costs and logistics losses, but also maximize user experience through differentiated appeasement, compensation, and refund strategies, thereby improving the platform's overall fulfillment control capabilities for high-complaint-risk categories (such as spray-painted flowers). Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram illustrating an application scenario provided in one embodiment of this application; Figure 2 This is a flowchart illustrating a method for aligning the fulfillment status and resending abnormal flower orders across platforms, as provided in an embodiment of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0028] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0029] Currently, the multi-platform data collaboration in the fresh flower e-commerce industry only enables process traceability, lacking a quantitative assessment mechanism for flower quality that runs through the entire fulfillment chain. It is impossible to quantify sensory deterioration caused by multiple factors, there are no differentiated handling solutions, and it is difficult to determine the responsibility for orders with abnormal quality.

[0030] Based on this, this application provides a method for aligning the fulfillment status and resending abnormal flower orders across platforms. It integrates two types of information sets: merchant fulfillment and delivery status. It analyzes multiple factors that cause losses in flower processing and transportation throughout the entire process, quantifies the limit of quality degradation of flowers upon arrival, and implements a tiered resending strategy for cross-platform flower orders. This solves the pain point of difficulty in quantifying and handling the deterioration of sprayed flowers, clarifies responsibilities, and improves cross-platform fulfillment alignment and user satisfaction.

[0031] Figure 1 This application provides an illustration of an application scenario. In the process of aligning the fulfillment status and resending abnormal orders for spray-painted floral materials across platforms, the method provided in this application establishes a correlation analysis between the user platform's order information set and the merchant platform's product information set. This quantifies the initial perception deviation of the floral materials under the influence of human processing and natural evolution. The introduction of the delivery platform's status information set further dynamically assesses the impact of the logistics environment on the flowers' ability to maintain their quality and the characteristics of subsequent quality changes. Ultimately, by integrating the user-merchant fulfillment deviation information set and the delivery risk assessment information set, the final delivery quality of the floral materials can be objectively predicted, and differentiated abnormal resending strategies can be implemented accordingly. This effectively solves the problem of chaotic resending caused by unclear responsibility definition in existing technologies, achieving intelligent alignment and abnormal resending control of fresh flower order fulfillment status.

[0032] Specifically, the method provided in this application is applied to any server, which interacts with the user platform, merchant platform, and delivery platform to obtain the user platform order information set provided by the user platform, the merchant platform product information set provided by the merchant platform, and the delivery platform status information set provided by the delivery platform. It objectively predicts the final delivery quality of the flowers, intelligently aligns the fulfillment status of flower orders and manages risks in advance, and outputs cross-platform flower order abnormal resending strategies to florists, thus realizing the quantitative assessment and hierarchical handling of flower order fulfillment status.

[0033] The specific implementation method can be referred to in the following embodiments, where the data mentioned in the embodiments are only for reference and examples, so that relevant personnel can better understand them.

[0034] Figure 2 This is a flowchart illustrating a method for aligning fulfillment status and resending abnormal flower orders across platforms, provided in one embodiment of this application. The method of this embodiment can be applied to servers in the above scenarios. Figure 2 As shown, the method includes: Example 1: S201. Obtain the user platform order information set and the merchant platform product information set. Based on the user platform order information set and combined with the merchant platform product information set, analyze the perceptual deviation characteristics of floral materials caused by human processing and natural evolution, as well as the changes of floral materials under physical combination and structural constraints, to obtain the user-merchant performance deviation information set.

[0035] The user platform's order information set can represent the attributes of the flower products selected by the user when placing an order at the terminal, including data on the type of bouquet and the quantity of flowers, with the user platform as the data source. The merchant platform's product information set can record the processing and storage parameters of flowers from raw materials to finished products, including flower processing information and flower storage information, with the merchant platform as the data source. After obtaining the above two types of information sets, the bouquet type identifier in the user order is matched with the processing records in the merchant database through a data interface. Based on the matching results, the perceptual bias characteristics of flowers caused by both human processing and natural evolution are analyzed. Among them, the bias caused by human processing stems from the physical changes in the size and shape of flowers caused by packaging methods, and the chemical effects of spraying on the color and aroma of flowers; the bias caused by natural evolution stems from the irreversible changes in the freshness and natural color of flowers caused by the temperature and humidity environment during storage. At the same time, the changes of flowers under physical combination and structural constraints are analyzed, namely the mutual compression and arrangement restrictions of multiple flowers in the same packaging space, and the resulting morphological distortions. The above analysis results are stored in a structured manner to construct a user-merchant performance deviation information set, which quantifies the difference between the initial quality state of the flowers before they leave the merchant and the user's expectations.

[0036] To ensure that records on the user platform, merchant platform, and delivery platform correspond to the same target order, the server standardizes the fields of records from each platform and constructs a cross-platform order mapping key based on the user platform order identifier, product SKU, order creation time, and delivery area code. This cross-platform order mapping key is determined according to the following formula: ; In the formula, K is the cross-platform order mapping key; Hash(·) is the hash mapping function; ID_u is the user platform order identifier; SKU is the bouquet product code; t_o is the order creation time; and Z_a is the delivery area code. The server converts the merchant's order acceptance time, outbound time, delivery order acceptance time, and estimated delivery time into a unified time base. Within a preset alignment time window, it selects the merchant record and delivery record with the same mapping key and the smallest time difference. When there are multiple candidate records, the record with the highest field completeness is selected first to form a cross-platform fulfillment status sequence for the same target order.

[0037] When aligning the user's desired state with the merchant's actual state, the server extracts the target flower quantity, target color, target bouquet width, target bouquet height, and target outline type from the product display images and order descriptions on the user platform. It also extracts the actual flower quantity, actual color, actual width, actual height, flower head center distribution, and damaged petal areas from the merchant's outbound front and side views. Taking width and height as examples, the dimensional deviation d_s is calculated using the following formula: ; In the formula, d_s represents the normalized shape and size deviation; W_a and H_a are the width and height of the actual bouquet in the merchant's outbound image, respectively; W_r and H_r are the baseline width and height corresponding to the order display image, respectively; clip{·,0,1} indicates that the calculation result is restricted to between 0 and 1. The actual size in the image is scaled using the known size of the packaging container or a calibration board taken with the image; the color, quantity, and outline words in the order description are mapped to standard fields and together with the image features constitute the user's expected state and the merchant's actual state.

[0038] S202. Obtain the delivery platform status information set. Based on the delivery platform status information set and the user-merchant fulfillment deviation information set, analyze the ability of flowers to maintain their quality during delivery and the characteristics of quality changes caused by environmental factors to obtain the delivery risk assessment information set.

[0039] The delivery platform status information set is a dataset describing the dynamic environmental parameters of flowers in the logistics chain, including delivery route information, delivery time, and delivery packaging type, with the delivery platform as the data source. This delivery platform status information set is correlated with the user-merchant fulfillment deviation information set generated in the preceding steps to analyze the flowers' ability to maintain their original shape and quality during delivery. This ability refers to the physical characteristics of the flowers in maintaining their original form and quality when subjected to external mechanical shocks and environmental influences. First, based on the road surface smoothness data in the delivery route information, combined with the initial morphological deviation recorded in the user-merchant fulfillment deviation information set, the degree of further damage to the size and shape of the flowers caused by transportation bumps is calculated. Subsequently, the characteristics of quality re-change caused by environmental factors are analyzed, comprehensively considering the temperature change trend during the delivery period and the impact of delivery packaging type (such as breathability and insulation) on the physiological activities of the flowers, assessing changes in respiration intensity, water loss rate, and possible chemical reactions (such as accelerated oxidation) during transportation. The physical deformation caused by transportation bumps is superimposed with the quality deterioration caused by environmental factors to obtain a delivery risk assessment information set, which predicts the degree of quality degradation of the flowers before they reach the user.

[0040] S203. Integrate user-merchant fulfillment deviation information set and delivery risk assessment information set, classify abnormal re-delivery levels, output and execute cross-platform flower order abnormal re-delivery strategies.

[0041] This system integrates user-merchant fulfillment deviation information and delivery risk assessment information, and uses a weighted fusion algorithm to calculate the expected quality status of flowers upon arrival at the user's location after processing, warehousing, and delivery. Based on this expected status, and against a preset quality grading standard, it classifies abnormal reshipment levels. Abnormal reshipment levels indicate the severity of order fulfillment abnormalities; for example, they can be categorized as Level 1 reshipments (discrepancies in sensory characteristics but still meeting basic delivery standards), Level 2 reshipments (damage to physical structural integrity), and Level 3 reshipments (loss of aesthetic value and presence of odor). After grading, a cross-platform flower order abnormal reshipment strategy is output and implemented. This strategy specifically includes: for Level 1 reshipments, a reassurance strategy of sending a description of the discrepancy and offering a coupon; for Level 2 reshipments, a compensation strategy of partial refund and reshipment of replacement flowers; and for Level 3 reshipments, a termination strategy of full refund and cancellation of the original order. This tiered processing mechanism allows for intervention in flower orders with different risk levels, reducing customer complaints and improving user experience.

[0042] Example 2: In some embodiments, the process of constructing a user-merchant performance deviation information set specifically includes: Step 1: The user platform order information set includes the type of bouquet and the quantity of flowers; Bouquet type can define the biological attributes and visual morphology of the flowers (e.g., roses, lilies, etc.), while flower quantity refers to the number of blossoms within the bouquet, used to quantify physical constraints such as packaging pressure and storage space occupancy. The merchant platform's product information set focuses on describing the objective processing history of the products during their circulation. By obtaining flower processing information, specific process parameters can be identified, such as the type of pigment, spraying pressure, and drying time in the coloring process. These parameters determine the physicochemical state of the flower surface. Flower storage information provides the environmental exposure history of the flowers before they leave the factory, including storage duration, environmental layout, and monitoring data.

[0043] By integrating the two types of information sets mentioned above, and aligning the user's expected standards with the merchant's actual processing status, an input data base covering the entire supply and demand side is constructed.

[0044] Step 2: Based on the type of bouquet and the quantity of flowers, combined with the flower processing information, analyze the changes in the size and shape of the flowers caused by the packaging method, as well as the impact of the spraying treatment on the color and smell of the flowers, to obtain the human-induced size and shape deviations and the spraying color and smell deviations.

[0045] Quantitative modeling is used to address physical and sensory changes caused by human intervention. For the analysis of how packaging methods affect the size and shape of flowers, the biomechanical properties of the flowers (such as petal fragility and stem compressive strength) are determined based on the type of bouquet. The filling density within the packaging volume is calculated based on the quantity of flowers. Then, the deformation effect of specific packaging methods (such as wrapping tightness and cardboard box compression structure) on the flowers under prolonged stacking or compression is analyzed, yielding the human-induced dimensional and morphological deviations. These deviations reflect the unnatural morphological shifts of the flowers under physical constraints.

[0046] This study focuses on analyzing the spraying parameters in floral material processing information to identify color and odor deviations. Spraying is a complex physicochemical process involving the bonding of pigments to the petal epidermis (waxy layer or velvety structure). By analyzing the impact of spraying on the color and odor of floral materials, the study assesses pigment adhesion and solvent evaporation characteristics. For example, the degree of pigment curing is calculated based on the number of sprays and drying time. Insufficient curing leads to decreased pigment adhesion, thus predicting potential color fading or color drift during subsequent transfers. Simultaneously, the evaporation rate and potential residual chemical odors at different temperatures are analyzed in conjunction with the type of solvent used (e.g., alcohols, esters). These chemical odors are then superimposed with the floral material's own fragrance to quantify the color and odor deviation caused by spraying. This deviation characterizes the sensory deviation introduced by artificial chemical treatment of floral materials.

[0047] Step 3: Based on the quantity of flowers, the type of bouquet, and the information on flower storage, analyze the effects of storage temperature and humidity on the freshness and natural color of the flowers to obtain the natural evolution color deviation.

[0048] This study isolates and quantifies the quality drift caused by natural environmental factors, eliminating human interference. Based on flower species, baseline values ​​for respiration intensity and transpiration rate are first determined for different flower types; for example, roses and lilies exhibit different sensitivities to temperature. Combining storage temperature and humidity data from flower storage information, the study analyzes how environmental parameters alter flower freshness. Specifically, by calculating the nutrient consumption rate caused by accelerated respiration due to high temperatures and the risk of fungal infection due to inhibited transpiration in high humidity environments, a freshness decay curve is generated. Simultaneously, regarding the impact of storage temperature and humidity on the natural color of flowers, the study analyzes biochemical processes such as chlorophyll degradation and inhibited anthocyanin synthesis to predict the degree of natural fading or darkening of color during the storage period, thus obtaining the natural evolutionary color deviation.

[0049] This bias provides a quantitative basis for distinguishing between natural aging caused by improper storage and human-caused damage caused by processing defects.

[0050] Step 4: Integrate human-caused size and shape deviations, spray color and odor deviations, and naturally evolved color deviations to construct a user-merchant fulfillment deviation information set.

[0051] A multi-source data fusion operation is performed, using human-induced size and shape deviations as feature vectors for the physical dimension, spray-painted color and odor deviations as feature vectors for the chemical sensory dimension, and naturally evolved color deviations as feature vectors for the biological metabolic dimension. Weight coefficients for each dimension are then configured based on the type of bouquet and the quantity of flowers. For example, the weight of human-induced size and shape deviations can be increased for fragile flowers; and the weight of spray-painted color and odor deviations is increased for products dominated by spray-painting processes.

[0052] To avoid arbitrarily adjusting the weights of each deviation based on subjective experience, the server reads the basic sensitivity coefficients of the target flowers to morphological damage, color treatment, and natural decay from the flower parameter database. Combined with process indicators such as whether the order was color-treated and whether the flowers are fragile, the server determines the normalized weights of each deviation and the initial fulfillment deviation between the user and merchant according to the following formula: ; ; In the formula, G is the weight normalization factor for the three deviation dimensions; i is the deviation dimension index, and the three deviation dimensions are human-induced size and shape deviation, spray color and odor deviation, and natural evolution deviation, respectively; w_i is the normalized weight of the i-th deviation dimension; s_i is the basic sensitivity coefficient of the target flower material to the i-th deviation dimension; λ_i is the process correction coefficient of the i-th deviation dimension; I_i is the corresponding process or category identifier, which is 1 when the corresponding feature is met, and 0 otherwise; D_0 is the user-merchant initial fulfillment deviation; M is the human-induced size and shape deviation; S is the spray color and odor deviation; N is the natural evolution deviation; w_m, w_s, and w_n are the weights corresponding to M, S, and N, respectively, and w_m+w_s+w_n=1. For unsprayed flowers, S is set to 0 and the remaining weights are renormalized; each basic sensitivity coefficient and process correction coefficient is obtained by calibrating historical samples of packaging, storage, and spraying of different flowers.

[0053] Example 3: In some embodiments, the process of constructing artificial dimensional and morphological deviations specifically includes: Step 1: Based on the quantity of flowers and the information on flower processing, analyze the physical compression characteristics of the external binding and wrapping methods on the original unfolded state of the flowers, as well as the physical restrictions of the internal space of the packaging container on the unfolding angle of the flowers, to obtain information on compression distortion.

[0054] The original extended state of the flowers refers to their natural growth form in an unconstrained natural environment, encompassing dimensions such as flower orientation, petal unfolding angle, and the natural curvature of the stem. The external binding operations, including the tightening force, binding position, and number of loops, as well as the wrapping methods involving the number of folds in the wrapping paper, the tightness of the wrapping, and the material rigidity, collectively constitute the physical pressure characteristics on the flowers. Simultaneously, the volumetric constraints of the packaging container force the flowers to converge their natural spreading angle within a limited range, resulting in adaptive changes in form. Compression distortion information is a quantitative description of the degree of morphological deviation caused by these external physical constraints.

[0055] Specifically, this information is obtained by establishing a mechanical model. First, the overall volume and cross-sectional area of ​​the bouquet are calculated based on the number of flowers. Combined with the binding force parameters and packaging container size parameters recorded in the flower processing information, the radial compressive force and axial compressive force on the bouquet are calculated. Then, based on Hooke's Law and the biomechanical properties of the flowers, the displacement and deformation of different parts of the flowers under external forces are simulated, generating compression distortion information that includes deformation displacement vectors, deformation area distribution, and deformation severity levels. For example, for dense bouquets with a large number of flowers (e.g., more than 24 stems), the compression distortion information can show that the stem bending angle in the middle is greater than 30 degrees, and the outer petals fold due to mutual compression. This process accurately captures unnatural morphological changes caused by packaging operations, providing a data foundation for subsequent assessment of its impact on ornamental value.

[0056] Step 2: Based on the type of bouquet and combined with information on compression distortion, analyze the differentiated tolerance characteristics of stem flexibility and flower fragility under packaging compression of different flowers to obtain compression tolerance information.

[0057] Introducing the biological differences among flower species aims to assess their inherent ability to resist compressive deformation. Different flower species possess distinctly different physical properties. For example, lily stems are thick and densely fibrous, exhibiting high flexibility and resistance to permanent bending under pressure; while tulip stems are hollow and thin-walled, easily breaking or flattening under lateral pressure. Similarly, flower vulnerability depends on petal thickness, texture, and connection method. Double-petaled peonies have numerous layers of petals that support each other, allowing them to maintain their shape under limited pressure, while single-petaled poppies have extremely thin petals that can break or detach even with slight pressure. These differentiated tolerance characteristics reflect the different physical responses exhibited by different flower species under the same compressive load.

[0058] The construction of compression tolerance information relies on a pre-established mapping database that stores the correspondence between flower types and their mechanical property parameters (such as stem bending stiffness and petal fracture toughness). Based on the bouquet type, the corresponding mechanical parameters of the target flower are retrieved from the database. Combined with the pressure values ​​and deformation directions from the compression distortion information, the stress distribution of the flower under the current compression state is calculated. The calculated stress values ​​are compared with the yield strength threshold of that flower type to determine whether the current deformation is elastic (recoverable) or plastic (unrecoverable), thus generating compression tolerance information that includes the probability of damage to each part, the proportion of irreversible deformation, and an overall tolerance level assessment. For example, when the compression distortion information indicates that the compression force on a certain part exceeds 2 Newtons, for varieties with poor stem flexibility, the compression tolerance information marks that part as having a high risk of breakage; while for varieties with high flexibility, it is only marked as having a slight risk of bending. This step realizes a shift from simple external force analysis to a comprehensive assessment considering the organism's own resistance capabilities, making predictions against human bias more consistent with biological reality.

[0059] Step 3: Integrate information on compression distortion and compression tolerance to construct artificial dimensional and morphological deviations.

[0060] The quantitative results of external physical stress are organically integrated with the assessment results of the flowers' inherent tolerance to stress, forming a comprehensive index to describe human-induced dimensional and morphological deviations. These deviations not only include visual changes in the geometric dimensions of the flowers but also implicitly contain information about quality degradation caused by physical damage. The integration process uses weighted logic to reflect the correlation between the two, rather than simply adding up the data.

[0061] In practice, a multi-dimensional feature fusion algorithm is employed to map geometric deformation parameters (such as dimensional compression rate and angular offset) from the compression distortion information and damage parameters (such as breakage risk coefficient and damaged area ratio) from the compression tolerance information into the same feature space. During the fusion process, weighting coefficients are set according to the impact of damage on ornamental value. For example, higher weights are assigned to flower vulnerability parameters that directly affect the integrity of the flower head, while relatively lower weights are assigned to slight bending of the stem. The final result is an artificial dimensional and morphological deviation that includes a comprehensive deviation score, annotation of key deviation locations, and deviation type classification (such as irreversible deformation-dominated or structural damage-dominated).

[0062] In one optional quantification method, packaging load, angular offset, petal breakage, and irreversible deformation probability are normalized to the same dimension, and human-caused dimensional deviations are calculated according to the following formula: ; In the formula, M represents the human-induced dimensional deviation; F_c is the equivalent compressive force calculated based on the binding force, packaging container size, and flower filling density; F_y is the target flower yield load threshold retrieved from the flower parameter library; θ is the actual deflection angle of the flower head or stem relative to the order reference direction; θ_y is the allowable deflection angle threshold of the target flower; A_d is the area of ​​folded, damaged, or obviously bent petals or stems in the outbound image; A_f is the area of ​​the effective image region of the flower; p_ir is the irreversible deformation probability determined based on the duration and proportion of stress exceeding the yield threshold; a_1, a_2, a_3, and a_4 are the weights of the corresponding components, all non-negative, and a_1+a_2+a_3+a_4=1. F_y and θ_y are stored separately according to the flower type, thus different deviation results can be obtained when the same packaging load is applied to different flowers.

[0063] Example 4: In some embodiments, the process of constructing the color odor deviation of the sprayed paint specifically includes: Step 1: Based on the type of bouquet and combined with the information on flower processing, analyze the differences in the adhesion ability of the surface texture of the petals of different flowers to the sprayed pigment, as well as the characteristics of the pigment covering the original natural color of the flowers, to obtain the color coverage deviation.

[0064] Color coverage deviation can be used to characterize the degree of deviation between the visual color of sprayed flowers and the user's expected color or the natural state of the flowers. The surface texture of petals, as the microscopic physical morphology of the petal epidermis, directly determines the adhesion ability of sprayed pigments. For example, varieties with a thick and smooth waxy surface (such as Ecuadorian roses) have lower surface energy, making it difficult for sprayed pigments (especially water-based pigments) to spread, resulting in weak adhesion and a tendency for droplet rolling. Conversely, varieties with abundant surface hairs or loose cellular structures allow pigments to penetrate and anchor more easily, resulting in relatively stronger adhesion. Parameters such as spraying pressure, pigment viscosity, and atomized particle diameter included in the flower processing information, along with the petal surface texture, jointly determine the final deposition form of the pigments. Poor pigment adhesion leads to insufficient coverage, resulting in exposed white areas; excessive pigment adhesion but an overly thick coating may obscure the original natural texture of the flowers.

[0065] When quantifying color coverage deviation, the occlusion characteristics of the original natural color of the flower material after pigment coverage are analyzed and calculated. This process is based on a color optical mixing model to evaluate the occlusion rate of the sprayed pigment layer on the underlying pigment layer of the flower material. For semi-transparent pigments, the additive color mixing effect between the pigment layer and the base color needs to be considered; for pigments with strong coverage, the uniformity of their complete occlusion of the base color is mainly evaluated. When the pigment coverage is uneven or there are missed areas, obvious color patches will appear, creating a visual deviation. The calculation results of color coverage deviation are characterized by indicators such as color difference value (ΔE) or coverage area ratio, which are used for subsequent risk assessment. For example, when the calculated color difference value ΔE exceeds a preset threshold (e.g., ΔE>5), color coverage deviation is determined to exist.

[0066] Specifically, the color difference and the area of ​​the missed spray are converted into a color coverage deviation in the range of 0 to 1, and calculated according to the following formula: ; In the formula, C_s is the color coverage deviation; ΔE is the color difference between the color of the flowers shipped by the merchant and the base color of the order in the CIE Lab color space; ΔE_lim is the allowable color difference threshold for the target product; A_u is the area of ​​missed spraying, color blocks or mottled areas; A_f is the area of ​​the effective image area of ​​the flowers; b_1 and b_2 are the weights of the color difference component and the uneven coverage component, respectively, b_1+b_2=1; min(·,1) means that when the color difference ratio is greater than 1, it takes 1.

[0067] Step 2: Based on the quantity of flowers and the information on flower processing, analyze the uniform spraying characteristics of the sprayed pigments on a batch of flowers, as well as the odor variation characteristics caused by the mixture of the residual odor after the solvent in the pigments evaporates and the fragrance of the flowers themselves, and obtain the odor mixing deviation.

[0068] Odor mixing deviation is used to quantify the sensory changes in sprayed floral materials at the olfactory level caused by solvent intervention. Sprayed pigments typically contain organic solvents (such as ethyl acetate, toluene, etc.) or aqueous solvent carriers, which do not immediately and completely dissipate after spraying. Based on the quantity of floral materials and combined with the spraying process parameters (such as spray gun movement speed, nozzle flow rate, and number of sprays) in the floral material processing information, the uniform spraying characteristics of sprayed pigments on batches of floral materials are analyzed. If the spraying is uneven, some floral materials may accumulate excessive solvent, resulting in excessively high local odor concentrations; and the stacking of floral materials during batch processing may also hinder solvent evaporation in the centrally located areas.

[0069] As solvents evaporate, residual chemical odor molecules are gradually released and physically mix or chemically react with the natural aroma of the flowers (such as linalool and citronellol, volatile organic compounds in floral fragrances), resulting in odor variations. This odor mixing process is not a simple additive process but a complex chemical interaction. For example, the pungent chemical odor of the solvent may mask the delicate fragrance of the flowers, creating an abnormal olfactory experience of an industrial smell mixed with a cloying sweetness. The sensory aversion resulting from this mixture is often higher than that of a single solvent or a single floral fragrance. By constructing an odor mixing model, calculating the ratio of solvent evaporation concentration to floral fragrance concentration, and combining this with the sensitivity weights of the human nose to different odors, an odor mixing deviation is derived. This deviation is represented by a dimensionless sensory aversion index. When this index exceeds a preset sensory threshold (for example, a judgment threshold can be set to 0.6, where 0 represents no odor and 1 represents intolerance), an odor mixing deviation risk is identified.

[0070] When a volatile organic compound (VOC) sensor is installed inside the delivery packaging, the residual VOC concentration is read directly. If no sensor is installed, the residual VOC concentration is estimated based on the coating amount, drying time, effective packaging volume, packaging permeability, and ambient temperature, and the odor mixing deviation is calculated using the following formula: ; In the formula, O_s represents the odor mixing deviation; C_v represents the concentration of residual volatiles in the packaging space as measured by the sensor or estimated by the model; C_ref represents the reference odor concentration of the corresponding spraying solvent; s_o represents the odor sensitivity coefficient of the target floral material and the order scenario; clip[·,0,1] indicates that the result is limited to between 0 and 1. The C_ref and s_o are calibrated using odor evaluation samples under different concentration conditions.

[0071] Step 3: Integrate color coverage deviation and odor mixing deviation to construct a spray color and odor deviation.

[0072] By fusing the obtained color coverage deviation and odor mixing deviation at the data level, a multi-dimensional spray color and odor deviation is formed. This integration process is based on a weighted evaluation strategy, rather than a simple numerical addition. For example, in some application scenarios, visual deviation may have a higher weight, while in scenarios that emphasize quality experience, olfactory deviation may have a veto power. The constructed spray color and odor deviation serves as a comprehensive evaluation indicator, fully reflecting the overall impact of the spraying process on the color and odor of the flowers. This deviation information set is configured to be output to the subsequent delivery risk assessment stage, serving as key data for determining whether the flowers meet the fulfillment standards and whether to trigger an abnormal re-delivery strategy.

[0073] Color and odor deviation of the sprayed paint shall be determined by the following formula: ; In the formula, S represents the color and odor deviation of the sprayed paint; C_s represents the color coverage deviation; O_s represents the odor mixing deviation; w_c and w_o are the weights of the color and odor components, respectively, and w_c + w_o = 1. For floral products that are primarily for visual display, increase w_c; for products with sealed packaging or sensitive odors, increase w_o. When O_s is not less than 0.80, the order is marked as a high-risk odor order, and the odor risk will no longer be offset by lower color deviation.

[0074] Example 5: In some embodiments, the process of constructing naturally evolved color deviation specifically includes: Step 1: Based on the quantity of flowers and combined with flower storage information, analyze the oxidation and discoloration characteristics of the flower cuts and epidermis due to contact with air to obtain oxidation and discoloration information.

[0075] The cut surface and epidermis of floral materials, as the interface between plant tissues and the environment after the protective layer is lost, undergo an enzymatic reaction with oxygen catalyzed by polyphenol oxidase after cutting due to the exposure of cell structures and the activation of physiological defense mechanisms, thus causing a color change. This oxidative discoloration process is affected by the ambient oxygen concentration, temperature, and the physiological state of the floral materials, and usually manifests as browning or blackening starting from the cut surface, which may spread towards the flower head with water transport.

[0076] Introducing the quantity of flowers as a biological variable into the analysis aims to consider the accelerated oxidation rate effect of limited gas exchange in the microenvironment under dense stacking conditions. Combining flower storage information, parameters such as oxygen concentration, ambient temperature, and airflow within the storage space are extracted. A pre-defined oxidation reaction kinetic model is used to simulate the color degradation trajectory of flowers due to continuous air contact during storage, thereby quantifying oxidation discoloration information. This information specifically includes data such as the proportion of browned area at the cut, the change in color difference value (ΔE) of the epidermis, and the propagation speed of the oxidation front. This information is constructed during the real-time monitoring phase after storage commences: multi-sensor data of the storage environment are collected, and a pre-defined oxidation rate calculation formula is used to numerically characterize the expected oxidation degree of each batch of flowers, forming a structured data record containing timestamps and spatial distribution information. By transforming microscopic biochemical changes into calculable engineering parameters, a quantitative basis is provided for subsequent deviation assessment.

[0077] The storage process is divided into n sampling time periods, and the corresponding decay rate κ_j and cumulative freshness loss L_f are calculated based on the temperature and humidity of each time period. ; ; ; In the formula, κ_j is the decay rate of the flower material in the j-th sampling time period; U_j is the cumulative decay amount up to the j-th sampling time period, U_0 is the initial cumulative decay amount, and U_n is the cumulative decay amount after all n sampling time periods have ended; k_0 is the decay coefficient of the target flower material under the reference temperature and humidity conditions; Q_10 is the respiration intensity correction coefficient for every 10 degrees Celsius increase in temperature; T_j is the storage temperature in the j-th sampling time period; T_0 is the reference temperature; γ is the humidity deviation correction coefficient; H_j is the storage relative humidity in the j-th sampling time period; H_0 is the suitable relative humidity of the target flower material; Δt_j is the duration of the j-th sampling time period; n is the number of sampling time periods; L_f is the cumulative freshness loss in the range of 0 to 1, and the larger the value, the more obvious the decrease in freshness. k_0, Q_10, T_0, γ, and H_0 are all retrieved from the preset parameter library according to the flower material type.

[0078] Step 2: Based on the information on oxidation and discoloration, and in conjunction with the type of bouquet, analyze the different sensitivity characteristics of different floral materials to the darkening of color and the distribution of spots caused by oxidation due to differences in the stability of petal pigments and the thickness of the epidermal wax layer, and obtain the natural evolution color deviation.

[0079] The stability of petal pigments reflects the ability of natural pigments such as anthocyanins and carotenoids to maintain their chemical bonds under different pH values, light, and oxidation conditions, determining the intrinsic strength of the flower's original color. The thickness of the epidermal wax layer constitutes a hydrophobic physical barrier, directly affecting the ease with which oxygen and water penetrate into epidermal cells, and is a key physical factor determining the visual manifestation of oxidation reactions. Due to differences in botanical classification, different flower species have different pigment chemical structures and epidermal micromorphologies. For example, the thicker wax layer of roses makes them exhibit a delayed oxidation effect, while flowers with thinner petals and less wax are more prone to obvious oxidation spots.

[0080] Based on the oxidation and discoloration information generated in the aforementioned steps, the current oxidation stage and degree of the floral materials are determined. In conjunction with the species characteristic database corresponding to the type of bouquet, the standard pigment stability parameters and epidermal wax layer thickness distribution data of this type of floral material are retrieved.

[0081] By comparing and analyzing different floral materials, we calculated their differential sensitivity characteristics under the same oxidation conditions. Specifically, we identified which floral materials tended to experience overall darkening of color under the same intensity of oxidation, and which tended to develop discrete spots in specific areas. A weighted algorithm was used to couple the physical barrier effect (wax layer thickness) with the chemical stabilization effect (pigment stability), ultimately outputting the natural evolution color deviation.

[0082] Natural evolutionary bias is further calculated using the following formula: ; In the formula, N represents the natural evolution deviation; L_f represents the cumulative freshness loss; ΔE_n represents the amount of color change caused by natural evolution during storage; ΔE_nlim represents the allowable natural color change threshold for the target flower material; R_m represents the mold risk determined based on the duration of humidity exceeding the limit, the area of ​​spots, and the fungal risk indicator; c_1, c_2, and c_3 are the weights of freshness, natural color difference, and mold risk, respectively, and c_1+c_2+c_3=1.

[0083] Example 6: In some embodiments, based on the delivery platform status information set and combined with the user-merchant fulfillment deviation information set, the ability of flowers to maintain their quality during delivery and the characteristics of quality re-change caused by environmental factors are analyzed to obtain a delivery risk assessment information set. This process specifically includes: Step 1: The delivery platform status information set includes delivery route information, delivery time period, and delivery packaging type.

[0084] By connecting to the delivery management system's database via a real-time interface, the system extracts and integrates delivery order details to form a delivery platform status information set. This includes specific information such as delivery route coordinates (starting and ending points), total mileage, road class (e.g., urban expressways, rural bumpy roads), and expected road conditions. The delivery timeframe is defined as the order fulfillment window, including the start shipping time, estimated delivery time, and weather forecast data for that period. The delivery packaging type records the specific packaging method used to carry the flowers, such as cardboard boxes, insulated bags, or polypropylene plastic boxes, along with the corresponding packaging material parameters.

[0085] Step 2: Based on the delivery route information and human-caused size and shape deviations, analyze the further impact of transportation bumps on the size and shape of the flowers to obtain transportation deformation information.

[0086] By utilizing road condition characteristics and vehicle dynamics parameters from delivery route information, the mechanical vibration environment during the delivery process is simulated. Specifically, based on road smoothness data and vehicle suspension system characteristics, the peak vertical acceleration, vibration frequency spectrum, and lateral impact force that may occur during transportation are calculated to quantify the intensity of transportation bumps.

[0087] The calculated turbulence intensity parameters are coupled with the compression distortion and crush tolerance information recorded in the human-caused dimensional and morphological deviation data for analysis. This process aims to evaluate the structural response of flowers subjected to external vibration on top of existing physical compression (from packaging). For example, for flower stems already under compression and compaction within the packaging, low-frequency, high-amplitude turbulence may cause fatigue bending; while for areas with pigment adhesion at the petal edges, high-frequency vibration may induce stress concentration, leading to pigment peeling or petal edge tearing. The analysis outputs data including the expected degree of deformation, damage location, and deformation probability distribution of the flowers at the end of transportation—that is, transportation deformation information.

[0088] Step 3: Based on the transportation deformation information, combined with the delivery time and natural color evolution deviation, analyze the impact of temperature changes during the delivery time on the quality of the flowers, and obtain information on quality changes during the time period.

[0089] Based on the delivery period, the corresponding time span and ambient temperature curve are obtained to calculate the temperature change rate, the duration of exposure to the highest temperature, and the temperature and humidity fluctuations throughout the entire process from shipment to receipt. Simultaneously, the color sensitivity characteristics of floral materials due to oxidation reactions are read from natural evolution color deviation data to determine the current response threshold of the floral materials to temperature changes.

[0090] This method overlays temperature change data with the heat sensitivity characteristics of floral materials. For example, for floral materials that have already shown initial oxidation and darkening during storage, if delivery occurs during the midday heat, the high temperature accelerates the transpiration of water within the petals and promotes the activity of polyphenol oxidase, causing the color to rapidly change from dark to brown. Furthermore, drastic temperature fluctuations can also cause changes in the permeability of petal cell membranes, further exacerbating color deterioration. By comprehensively evaluating the synergistic effects of heat aging, dehydration, and oxidation, data characterizing the rate of quality degradation, final freshness retention, and degree of color deterioration of floral materials within a specific time period—that is, time-bound quality change information—is generated.

[0091] Step 4: Based on the information on quality changes over time, combined with the type of delivery packaging and the deviation of spray color and odor, analyze the impact of the packaging environment on the color and odor of the flowers to obtain a delivery risk assessment information set.

[0092] This study analyzes the physical properties corresponding to different types of delivery packaging, focusing on extracting the light transmittance and air permeability of the packaging materials. For light transmittance, it analyzes the transmittance of the packaging material to ultraviolet and infrared light in the solar spectrum, assessing the risk of photodegradation and aging of floral pigments (especially sprayed and natural pigments) due to light intensity. For air permeability, it measures the gas exchange rate inside and outside the packaging to determine the relative airtightness of the internal microenvironment.

[0093] By combining the odor mixing deviation and color coverage deviation recorded in the color and odor deviation data of the sprayed paint, a packaging microenvironment model was constructed. In highly sealed packaging (such as polypropylene plastic boxes), due to poor air permeability, volatile organic solvents (such as ethyl acetate and toluene) in the sprayed paint cannot diffuse in time, leading to their accumulation within the packaging space. High concentrations of solvent molecules not only further dissolve or damage the pigment layer on the surface of the floral materials, causing chemical fading or spotting and peeling of the color, but also react with volatile aroma molecules released by the flowers themselves, generating irritating or unpleasant mixed gases. Furthermore, if the packaging is too transparent, light exposure may accelerate solvent evaporation and photo-oxidative decomposition of the pigments, further worsening the odor environment.

[0094] By quantifying the cumulative solvent concentration within the packaging, the sensory aversion index after gas mixing, and the destructive coefficient of light on color stability, and integrating the deformation and quality degradation data obtained from the aforementioned steps, a comprehensive assessment result is finally formed, which includes physical damage risk, chemical deterioration risk, and sensory experience risk—that is, the delivery risk assessment information set.

[0095] The road vibration characteristics provided by the above delivery route information and the temperature field environmental characteristics provided by the delivery period are used as the input boundaries of the physical force field and the environmental chemical field, respectively.

[0096] Example 7: In some embodiments, the process of constructing transport deformation information specifically includes: Step 1: Based on the delivery route information, analyze the vertical vibration frequency characteristics caused by differences in road surface smoothness along the transportation route, as well as the horizontal inertial characteristics during vehicle start-stop and turning processes, to obtain bump and impact information.

[0097] Road condition data for the transportation route is extracted from the acquired delivery route information. This delivery route information is not simply a set of route coordinates, but a structured sequence containing road surface grade, paving quality information, and driving behavior data, reflecting the dynamic excitation characteristics of the transportation vehicle in spatial and temporal dimensions. Based on the above road condition data, road surface smoothness differences are analyzed. Road surface smoothness differences, as a comprehensive measure of the macroscopic texture and microscopic unevenness of the road surface, directly determine the excitation intensity generated during vehicle operation. For road surfaces with different smoothness, the suspension system and tires of the transportation vehicle will produce different spectral responses, thus forming vertical vibration frequency characteristics. These vertical vibration frequency characteristics characterize the periodic or random vibration energy distribution experienced by the packaging container and the flowers inside in the vertical direction, and are a key external input leading to fatigue damage to the flowers.

[0098] By combining driving behavior data during the delivery process, we further analyzed the horizontal inertial characteristics during vehicle start-stop and steering. Vehicle start-stop (acceleration and braking) and steering alter the vehicle's motion state, resulting in inertial forces acting on the flowers in the horizontal direction. Unlike the periodicity of vertical vibration, this horizontal inertial characteristic typically manifests as low-frequency, high-amplitude transient impacts, which can easily disrupt the static equilibrium of the flowers within the packaging. Using a vehicle kinematic model, acceleration data was converted into inertial force vectors, thereby quantifying the degree of horizontal disturbance to the flowers. By combining the vertical vibration frequency characteristics and the horizontal inertial characteristics, bump impact information was constructed. This bump impact information, as a quantitative indicator describing the mechanical load of the transportation environment, includes not only the amplitude, frequency, and duration of vibration but also the direction and peak energy of the inertial impact, providing an external force input boundary for subsequent assessment of physical damage to the flowers.

[0099] The server divides the triaxial acceleration data during the delivery process into n sampling time periods and calculates the normalized turbulence intensity B according to the following formula: ; ; ; In the formula, E_j is the equivalent vibration energy of the j-th sampling time period; V_j is the accumulated equivalent vibration energy up to the j-th sampling time period, V_0 is the initial accumulated value, and V_n is the accumulated value after all n sampling time periods have ended; a_zj is the vertical acceleration of the j-th sampling time period; a_xj and a_yj are the accelerations in the two horizontal directions, respectively; a_ztol and a_htol are the vertical acceleration tolerance threshold and horizontal acceleration tolerance threshold of the target flower material, respectively; η is the horizontal inertia influence correction coefficient; Δt_j is the duration of the j-th sampling time period; T_d is the total delivery time; n is the number of sampling time periods; and B is the normalized turbulence intensity. The acceleration tolerance threshold is retrieved from the parameter library based on the flower material type and packaging fixation method.

[0100] Step 2: Based on the information about the bumps and impacts, combined with the human-caused size and shape deviations, analyze the characteristics of the flower head displacement caused by the impact of external force on the basis of the original physical compression, and obtain the information on the distortion of the bouquet.

[0101] The aforementioned bump and impact information is used as an external excitation source and coupled with the artificial dimensional and morphological deviations generated in the previous steps for analysis. These artificial dimensional and morphological deviations reflect the initial physical compression and morphological constraints on the flowers during the packaging stage due to bundling and wrapping. The flowers are not rigidly fixed inside the packaging container but are in a constrained equilibrium state. When the bump and impact information during transportation exceeds a certain threshold, the original equilibrium state is disrupted.

[0102] The flower head offset characteristics are analyzed under the combined effects of pre-existing physical pressure and external impact. Pre-existing physical pressure refers to the static constraint force exerted on the target flower by the inner wall of the packaging container, lining paper, or other floral materials; external impact refers to the dynamic load quantified by information from bumpy impacts. As a part with a relatively concentrated center of mass, the flower head is highly susceptible to displacement under the aforementioned mechanical conditions. Through dynamic simulation or empirical models, the angle and displacement of the flower head from its initial equilibrium position under different inertial forces and vibration accelerations are calculated, thus determining the flower head offset characteristics. These characteristics describe not only the magnitude of the displacement but also the direction of the displacement (e.g., tilting towards one side of the packaging) and whether irreversible plastic deformation occurs.

[0103] By integrating the above features, bouquet distortion information is constructed. This bouquet distortion information describes the degree of damage to the overall shape of the entire bouquet, going beyond the displacement of individual flowers. It characterizes the degree of distortion in the overall geometric structure of the bouquet, such as whether there is overall skewing or scattered flower heads. The generation of this information transforms the abstract concept of transportation bumps into specific structural distortion indicators, providing an intuitive morphological basis for subsequent judgments on whether relocation is necessary.

[0104] Step 3: Based on the information on the distortion of the bouquet, analyze the characteristics of mutual compression and friction between individual flower materials caused by displacement, as well as the tendency of petal edges to curl and fall off due to collision, to obtain the transportation deformation information.

[0105] Based on the aforementioned information on bouquet distortion, a deeper analysis of the physical damage at the microscopic level is conducted. Bouquet distortion signifies a change in the relative positions of individual flower pieces, directly leading to mutual compression and friction. Under normal packaging conditions, flowers typically maintain a certain gap or gentle contact; however, after bouquet distortion occurs, previously separate flower stems may overlap, with flower heads in close contact. Under continuous external force, this contact generates intense compression and friction.

[0106] This study focuses on the microscopic damage caused by compression and friction, specifically analyzing the curling and shedding trends of petal edges due to collisions. Petal edges are typically the most vulnerable parts of flowers, most prone to moisture loss, and have the lowest mechanical strength. During collisions, stress concentration easily occurs at the petal edges, leading to irreversible curling deformation and even brittle fracture and shedding. By simulating the response of petal materials under contact stress, the degree of curling and the probability of shedding are quantified, resulting in specific damage assessment indicators.

[0107] Finally, by integrating the characteristics of mutual compression and friction, petal edge curling, and shedding trends, transportation deformation information is generated. This transportation deformation information is a comprehensive summary of the physical damage suffered by the flowers during the delivery stage, covering the entire chain from macroscopic structural distortion to microscopic tissue damage. This information directly reflects the physical appearance quality of the flowers when the user receives them and is the core data support for determining whether order fulfillment is satisfactory and what level of anomaly relocation strategy to trigger. Through this refined deformation information construction, the negative physical impact of the transportation process on the quality of the flowers can be accurately captured, thereby achieving anomaly control.

[0108] The transport deformation risk D_t is calculated according to the following formula: ; In the formula, D_t represents the transportation deformation risk; B represents the normalized turbulence intensity; d_h represents the predicted offset of the flower head relative to its initial position; d_lim represents the allowable offset of the flower head in the target flower material; A_p represents the area of ​​the petal curling, collision damage, or detachment region; A_f represents the area of ​​the effective image region of the flower material; d_1, d_2, and d_3 are the weights of the turbulence, flower head offset, and petal damage components, respectively, and d_1+d_2+d_3=1.

[0109] The aforementioned technical process couples macroscopic road condition stimuli with microscopic physical damage to the flowers. The unevenness of the road surface and the bumps and impacts generated by driving behavior act on the flowers that already have human-induced shape deviations, causing flower head displacement and overall distortion. This leads to internal compression friction and edge collisions, ultimately forming transport deformation information that can be quantified and assessed. This collaborative process makes the previously unpredictable physical losses transparent and controllable, providing a basis for aligning the fulfillment status of cross-platform flower orders.

[0110] Example 8: In some embodiments, the process of constructing time-period quality change information specifically includes: Step 1: Based on the transportation deformation information, analyze the damage to the surface of the compressed parts of the flower material and the water loss characteristics caused by the obstruction of the internal water supply channels of the flower material to obtain the flower material damage information.

[0111] Transportation deformation information, used as input data, records the physical structural changes of flowers after experiencing bumps and jostling during the preceding delivery process. By analyzing this information, the locations of deformation caused by external mechanical pressure are first identified. For these compressed areas, the state of damage to their microstructure is analyzed in depth, with a focus on assessing the degree of rupture in the cuticle and epidermal cells, i.e., epidermal damage. This analysis aims to quantify the degree of loss of epidermal barrier function, as the integrity of the epidermis directly affects the flower's water retention capacity. Simultaneously, the structural impact of physical compression on the internal vascular bundle system of the flower is examined. As the main channels for transporting water and inorganic salts, the vascular bundles may flatten or twist under continuous physical pressure, leading to obstructed water transport pathways, i.e., blocked water supply channels. Based on this physical morphological analysis of the internal vascular structure, the decrease in cell turgor pressure and water deficit caused by reduced water transport efficiency are further evaluated, thus obtaining water loss characteristics. By integrating the external barrier failure indicator of epidermal damage and the internal physiological deficiency indicator of water loss, a damage information system for floral materials was constructed, encompassing dimensions such as damage location, damage area, and water loss rate. This damage information not only reflects the current state of physical damage but also provides a crucial data foundation for subsequent assessment of the accelerated decline in physiological functions.

[0112] Step 2: Based on the information about damaged flowers and the delivery time, analyze the characteristics of changes in the respiration intensity of flowers caused by changes in ambient temperature during delivery, as well as the characteristics of accelerated water loss in damaged areas, to obtain information on physiological depletion.

[0113] Using the information on damaged flowers obtained in the preceding steps, and combining it with the environmental parameters corresponding to the delivery time slots provided by the delivery platform, a dynamic physiological state assessment was conducted. The delivery time slots implicitly contain information about changes in ambient temperature. For example, delivery during the midday heat can alter the metabolic level of the flowers due to increased temperatures in the delivery vehicle or the external environment. First, the dynamic fluctuations in the intensity of respiration in the flowers, i.e., the characteristics of respiration intensity changes, were analyzed under the influence of ambient temperature changes. Specifically, as the ambient temperature rises, the enzyme activity of the flowers increases, and the respiration rate accelerates, leading to the accelerated consumption of stored substances such as starch within the cells. While this process maintains life activities, it also accelerates the aging process of the flowers. Next, the coupling effect of physical damage and environmental stress was analyzed in detail. For the epidermal damage areas indicated in the flower damage information, the loss of the cuticle layer exposes the water directly to the surrounding environment. When the ambient temperature rises, the relative humidity of the air usually decreases, and the saturated vapor pressure difference increases, causing the water evaporation rate of the damaged areas to be much higher than that of healthy areas, exhibiting a characteristic of accelerated water loss. By combining the leakage effect caused by physical defects with the energy consumption effect caused by temperature increases, the fresh weight loss rate and vitality decline rate of flowers per unit time are quantitatively assessed, thus obtaining physiological loss information. This physiological loss information characterizes the overall vitality consumption level of flowers during a specific delivery period and is an important basis for judging whether irreversible quality degradation has occurred in flowers.

[0114] Step 3: Based on physiological wear and tear information and natural evolutionary color deviation, analyze the color deepening phenomenon of damaged parts on the basis of the original oxidation and discoloration, as well as the color darkening phenomenon of healthy parts due to water loss, to obtain information on quality changes over time.

[0115] Based on the physiological deterioration information generated in the aforementioned steps, and incorporating natural evolutionary color deviation, the final visual quality of floral materials at the delivery terminal is predicted. Natural evolutionary color deviation records the initial color changes of floral materials during storage due to oxidation reactions, such as browning at the cut or slight darkening of petal edges. First, the color evolution of damaged areas on top of the original oxidative discoloration is analyzed. Due to increased cell membrane permeability in damaged areas, oxidases such as polyphenol oxidases are more likely to come into contact with substrates. Combined with metabolic disorders indicated by physiological deterioration information, this leads to abnormally active oxidation reactions in these areas, resulting in a deepening of color on top of the original oxidative discoloration, such as slight browning developing into dark brown spots or patches. Simultaneously, the quality changes in healthy areas are also analyzed. Although healthy areas do not experience physical damage, physiological deterioration information reveals an overall trend of moisture loss. As the intracellular water content of healthy areas decreases, the concentration of cell sap increases and the refractive index of light changes. Combined with the degradation of pigments such as anthocyanins, healthy areas exhibit a darkening of color, losing their original bright color and luster. By combining the color deepening of damaged areas with the color darkening of healthy areas, a time-series quality change information describing the overall visual morphology of floral materials was constructed. This time-series quality change information not only covers specific color parameter changes (such as hue shift and reduced brightness), but also relates to the distribution of color spots caused by damage, thus providing an intuitive and quantitative basis for subsequent abnormal relocation decisions.

[0116] After normalizing the loss of freshness, moisture loss, color deterioration, and damage to the skin or water-bearing tissues during the delivery period, the risk of quality change during the period is calculated using the following formula: ; In the formula, Q_t represents the risk of quality change during the time period; L_t represents the freshness loss calculated based on the temperature and humidity curve during the delivery period; W_l represents the moisture loss rate relative to the initial fresh weight at shipment; C_d represents the normalized color degradation value formed by the darkening of the color of damaged parts and the darkening of the color of healthy parts; P_s represents the combined value of the proportion of the area of ​​skin damage and the proportion of the water channel obstruction; e_1, e_2, e_3 and e_4 are the corresponding component weights, all of which are non-negative and e_1+e_2+e_3+e_4=1.

[0117] Example 9: In some embodiments, based on time-varying quality change information, combined with delivery packaging type and color and odor deviation of spray paint, the impact of the packaging environment on the color and odor of the flowers is analyzed to obtain a delivery risk assessment information set. This process specifically includes: Step 1: Based on the information on quality changes over time, combined with the type of delivery packaging, analyze the limiting characteristics of the light transmittance and air permeability of the packaging material on the maintenance of the flower color, as well as the shielding or mixing effect of the packaging material's own adsorption or release properties on the flower's odor, and obtain information on the interaction effects.

[0118] The time-varying quality change information generated in the preceding steps is used as the basic input. This information already includes changes in color and physiological state of the flowers due to damage and water loss during transportation. Based on this, the type of delivery packaging is introduced as a key variable to analyze the secondary shaping effect of the packaging microenvironment on the quality of the flowers.

[0119] The light transmittance of packaging materials is a crucial parameter affecting color retention. For spray-painted floral materials, excessively strong or specific light spectra can accelerate the photolysis of organic components in the pigments. Light transmittance depends not only on the material's transparency but also on its ability to filter ultraviolet light. For example, if fully transparent polypropylene packaging is used, direct sunlight exposure to ultraviolet rays may cause photochemical degradation of the pigment surface, resulting in a chalky fading of the originally vibrant spray-painted surface. Conversely, using kraft paper with added ultraviolet absorbers or dark, opaque packaging can effectively block the photo-oxidation reaction pathway, protecting the color. By comparing the spectral transmittance of packaging materials with the photosensitivity characteristics of the spray-painted pigments, the degree to which packaging limits or protects color retention can be quantified.

[0120] Meanwhile, the air permeability of packaging materials directly determines the gas exchange rate of the internal microenvironment. For damaged flowers or those in a period of vigorous respiration, appropriate air permeability helps dissipate heat and ethylene generated by respiration, preventing high temperatures from accelerating flower aging. However, for spray-painted flowers, excessive air permeability may accelerate the evaporation of residual solvents. While this reduces the accumulation of solvent odor, it may also lead to localized temperature differences caused by the latent heat carried away by solvent evaporation, thus affecting the stability of the pigment adhesion interface. Air permeability is usually characterized by the oxygen and water vapor transmission rates of the material. These parameters are used to assess the steady-state maintenance ability of the internal temperature and humidity field of the packaging, thereby determining its dynamic limiting characteristics for color maintenance.

[0121] Furthermore, the adsorption or release properties of packaging materials themselves constitute a microenvironmental regulation mechanism for odor. Certain porous materials (such as corrugated cardboard) possess strong physical adsorption capabilities, acting as odor sponges to adsorb volatile organic compounds released by floral materials and residual solvent molecules from sprayed pigments, thus shielding or diluting the original natural aroma or odor of the flowers. Conversely, some plastic films (such as polyvinyl chloride, PVC) may have plasticizers or stabilizers added during processing. These chemical additives may be slowly released in a sealed environment or when heated, mixing with the aroma molecules emitted by the flowers and causing adulteration, resulting in the originally pure floral fragrance being masked by a plastic or chemical smell. By establishing material adsorption-desorption models and additive release models, the physicochemical effects of packaging materials on odor are quantified, thereby generating information on the interactive influence of color maintenance limiting characteristics and odor regulation characteristics. This information reflects that packaging is not a passive container, but an active factor actively involved in the evolution of floral quality.

[0122] Step 2: Based on the interaction information and the color and odor deviation of the sprayed pigment, analyze the secondary mixing and crosstalk characteristics of the residual solvent volatiles of the sprayed pigment in the packaging space and the natural fragrance of the flowers to obtain the mixed crosstalk information.

[0123] Based on the aforementioned interactive information, and combined with the pre-obtained color and odor deviations in the spraying process, an in-depth analysis of the complex aerosol chemical environment within the packaging space was conducted. The color and odor deviations in the spraying process quantify the differences in color coverage and the initial odor mixing state caused by the spraying process itself, while the airtight or semi-airtight nature of the packaging space provides a site for these volatile substances to undergo secondary reactions.

[0124] Specifically, residual solvent volatiles (such as ethyl acetate and toluene) from the sprayed pigments accumulate continuously within the confined space of the packaging. When the packaging material has low permeability, these volatiles cannot dissipate in time, leading to an increase in local partial pressure. High concentrations of solvent molecules coexist at the microscopic level with the aroma components released by the flowers themselves (such as linalool, citronellol, and other terpenoids). The two are not simply physically superimposed, but may undergo a secondary mixing reaction. This mixing may manifest as competition or masking at the sensory level, where a strong, pungent solvent odor dominates at the olfactory receptor level, suppressing the pleasant perception of floral fragrance; it may also manifest as intermolecular interactions at the chemical level, where solvent molecules act as undesirable solvents for certain aroma components, altering the volatility profile of aroma molecules and causing the originally fresh floral fragrance to become unpleasant.

[0125] Crosstalk characteristics refer to the dynamic cyclical influence among packaging materials, solvent volatiles, and floral aromas. For example, if packaging materials adsorb solvents, they may temporarily reduce solvent concentration and mitigate odor impact; however, as the temperature inside the packaging rises (e.g., due to delivery vehicles being exposed to direct sunlight), the material may desorb, releasing the previously adsorbed solvents and causing secondary odor release; or, additive molecules released from the packaging materials may synergistically interact with solvent molecules, generating new, unpleasant mixed odor molecules. By analyzing the mass transport dynamics and sensory psychology characteristics within the packaging space, the odor intensity change curves and sensory pleasure reduction index of solvent volatiles and natural floral aromas after mixing are calculated to identify this crosstalk characteristic. For example, for solvent-based sprayed roses, in sealed plastic packaging, the mixing of ethyl acetate and rose aroma may lead to an increased sensory score for a chemical smell, and this crosstalk causes an olfactory impact due to the sudden drop in concentration upon opening the box. Based on this analysis, mixed crosstalk information is generated, which quantifies the degree to which the packaging environment modifies or distorts the final sensory experience of the sprayed flowers.

[0126] Between adjacent sampling times, the concentration of residual volatiles in the packaging space is iteratively updated according to the following formula: ; In the formula, C_v(t) and C_v(t+Δt) are the concentrations of residual volatiles in the packaging space at time t and time t+Δt, respectively; q_v is the release rate of residual volatiles per unit of floral material at the current temperature; N_f is the number of floral materials in the packaging; V_p is the effective volume inside the packaging; k_a is the gas exchange coefficient determined based on the oxygen permeability, opening area, and number of gas exchanges in the packaging; and Δt is the time interval between adjacent sampling times.

[0127] Step 3: Integrate interactive impact information and mixed crosstalk information to construct a delivery risk assessment information set.

[0128] By logically fusing interactive impact information with mixed crosstalk information, a final delivery risk assessment information set is constructed to guide decision-making. This step summarizes and refines the aforementioned multi-dimensional risk factors such as physical deformation, physiological wear and tear, color changes, and odor mixing.

[0129] Interactive impact information focuses on the effects of packaging materials, acting as physical barriers and chemical media, on floral materials, such as the risk of light fading due to light transmittance and the risk of odor dilution or concentration due to adsorption. Mixed crosstalk information focuses on the dynamic interactions between various volatile components within the packaging microenvironment and their resulting sensory consequences. The combination of these two types of information constitutes a complete risk profile of the complex system of packaging-floral materials-environment.

[0130] A weighted fusion strategy was adopted when constructing the risk assessment information set. For example, for high-end spray-painted floral products with extremely high odor sensitivity, the odor intensity index in the mixed crosstalk information will be given a higher risk weight; while for floral arrangements that emphasize color visual display, color maintenance limitation characteristics in the interaction impact information (such as color change due to light transmission) will become the core assessment indicator. Through cross-validation and complementarity of multi-source information, the one-sidedness of single-dimensional assessment can be avoided. For example, if the risk of physical deformation is low, but the risk of mixed crosstalk is extremely high (such as severe odor accumulation due to airtight packaging), the comprehensive risk assessment will still determine that the order has a high fulfillment risk. The final delivery risk assessment information set includes the prediction of the overall quality status of the flowers at the end of delivery, the probability distribution of potential user complaint points (such as pungent smell, dark color), and the quantitative data foundation to support the formulation of subsequent abnormal re-delivery strategies.

[0131] After normalizing the odor, light fading, and adsorption-desorption crosstalk within the packaging, the packaging microenvironment risk P_t is calculated according to the following formula, and further integrated with the transportation deformation risk and the quality change risk over time: ; ; In the formula, P_t represents the packaging microenvironment risk; R_odor represents the odor risk determined by the concentration of residual volatiles and the odor sensitivity coefficient; R_light represents the light fading risk determined by the packaging transmittance, light intensity, and pigment photosensitivity; R_ads represents the crosstalk risk caused by the adsorption, desorption, or release of additives from packaging materials; f_1, f_2, and f_3 are the corresponding component weights, and f_1+f_2+f_3=1; R_d represents the delivery risk; D_t represents the transportation deformation risk; Q_t represents the time-period quality change risk; g_1, g_2, and g_3 are the weights of transportation deformation, time-period quality change, and packaging microenvironment risk, respectively, and g_1+g_2+g_3=1; μ is the synergistic amplification coefficient of physical damage and environmental degradation; clip(·,0,1) indicates that the result is restricted to between 0 and 1. The product term D_tQ_t is used to reflect the synergistic effect of accelerated deterioration of physically damaged floral materials under high temperature and dehydration conditions.

[0132] Example 10: In some embodiments, user-merchant fulfillment deviation information sets and delivery risk assessment information sets are integrated to classify abnormal re-delivery levels, and a cross-platform flower order abnormal re-delivery strategy is output and executed. This process specifically includes: Step 1: Based on the user-merchant fulfillment deviation information set and combined with the delivery risk assessment information set, analyze the expected quality degradation limit of the flowers when they arrive at the user under the superposition of the transportation deformation information, the time period quality change information and the interaction impact information, and obtain comprehensive fulfillment risk information.

[0133] Utilizing multi-source data fusion technology, the user-merchant fulfillment deviation information set (including processing and warehousing deviations) generated in previous steps and the delivery risk assessment information set (including transportation deformation information, time-lapse quality change information, and interaction impact information) are used as input variables. A weighted overlay algorithm is employed to quantitatively assess the final state of the flowers after undergoing the entire supply chain. The expected quality degradation limit is not a single-dimensional numerical decrease, but a comprehensive quantitative representation of the overall ornamental value of the flowers. It reflects the worst-case quality boundary that the flowers may exhibit upon arrival at the user's end after a series of continuous deterioration processes, including physical compression, pigment loss, oxidation discoloration, and solvent evaporation crosstalk, starting from their initial shipping state. Based on the comprehensive fulfillment risk information output by this assessment process, multi-dimensional indicators such as the flowers' physical morphology retention, color conformity, and odor acceptability are aggregated, providing a quantitative basis for subsequent decision-making. This overlay analysis mechanism of end-to-end risk can identify unacceptable fulfillment risks caused by the amplification of minor deviations during the delivery process.

[0134] The overall performance risk R is calculated using the following formula: ; In the formula, R represents the overall fulfillment risk; D_0 represents the initial fulfillment deviation between the user and the merchant; R_d represents the delivery risk; h_1 and h_2 are the basic weights of the initial fulfillment deviation and the delivery risk, respectively, h_1+h_2=1; λ is the coupling coefficient of the amplification of initial defects during the delivery stage; D_0R_d is the interaction term between initial defects and delivery risk; clip(·,0,1) indicates that the overall fulfillment risk is limited to between 0 and 1. h_1, h_2, and λ are calibrated based on the arrival quality inspection results of historical orders.

[0135] Step 2: Based on the comprehensive performance risk information, classify the corresponding abnormal reassignment levels, output and execute cross-platform flower order abnormal reassignment strategies.

[0136] Based on preset risk threshold ranges, comprehensive performance risk information is mapped to different abnormal resale levels, establishing a correlation between technical assessment results and commercial handling strategies. The classification of abnormal resale levels strictly follows the severity of the risk, specifically covering Level 1 resale (flowers with sensory differences but still meeting basic delivery standards), Level 2 resale (flowers with damaged physical structural integrity), and Level 3 resale (flowers that have lost their ornamental value and have an odor). After determining the level, the corresponding strategy is automatically invoked, outputting and executing cross-platform abnormal resale strategies for flower orders, achieving an automated closed loop from risk identification to handling execution.

[0137] In an optional calibration implementation, when 0 ≤ R < 0.25, it is determined as normal fulfillment, maintaining the original order fulfillment milestone; when 0.25 ≤ R < 0.50, it is determined as Level 1 reshipment; when 0.50 ≤ R < 0.75, it is determined as Level 2 reshipment; and when 0.75 ≤ R ≤ 1, it is determined as Level 3 reshipment. The above intervals are calibrated based on the arrival quality inspection labels of historical orders, user complaint results, and re-delivery results, and can be configured separately for different flower material categories.

[0138] To prevent the overall average result from masking a single serious defect, hard trigger conditions are set: when the irreversible deformation probability p_ir is not less than 0.30, or when flower head breakage or stem breakage is detected, the anomaly level is not lower than level two; when the odor mixing deviation O_s is not less than 0.80, or the freshness retention rate upon arrival is not greater than 0.25, or when moldy areas are detected, it is directly judged as level three. When both the hard trigger condition and the overall performance risk range are met, the higher-level judgment result is adopted.

[0139] When it is determined that a change of fulfillment node is needed, the server obtains the inventory, product matching degree, and order availability status of candidate merchants from the merchant platform, and the estimated delivery time from the candidate merchant to the user and the risk of the candidate route from the delivery platform, and calculates the re-delivery matching score of the j-th candidate merchant according to the following formula: ; In the formula, S_j is the reassignment matching score of the j-th candidate merchant; I_j is the inventory satisfaction of the candidate merchant; Q_j is the matching degree between the candidate product and the original order in terms of flower type, quantity, color, and packaging method; t_j is the estimated delivery time corresponding to the candidate merchant; t_max is the maximum allowed remaining fulfillment time for the order; R_j is the predicted risk of the candidate delivery route; r_1, r_2, r_3, and r_4 are the weights of the corresponding components, and r_1+r_2+r_3+r_4=1. The server selects the candidate merchant with sufficient inventory, estimated delivery time not exceeding t_max, candidate route risk below the preset upper limit, and the largest S_j as the alternative fulfillment node.

[0140] Step 3: Abnormal reassignment levels include Level 1 reassignment for flowers with sensory differences but still meeting basic delivery standards, Level 2 reassignment for flowers with damaged physical structural integrity, and Level 3 reassignment for flowers that have lost their ornamental value and have an odor.

[0141] The above grading standards are specifically defined as follows: Level 1 grading mainly applies to situations where the flowers only have minor color differences or the bouquet shape is slightly shrunken, which are non-structural problems. In this case, the flowers still possess the basic attributes of a commodity and meet the delivery standards. Level 2 grading applies to situations where the flowers have suffered irreversible physical damage, such as flower heads breaking due to severe shaking, large areas of petals falling off, or stems being severely bent due to compression. Such damage directly destroys the ornamental integrity of the flowers. Level 3 grading is for extreme situations where the flowers are severely yellowed or wilted, or where the paint solvents accumulated in the sealed packaging produce a pungent chemical odor, or the flowers themselves rot and produce a foul smell. In this case, the product completely loses its usability and the possibility of delivery.

[0142] Step 4: When the issue is determined to be a Level 1 reassignment, implement a reassurance strategy by sending the user an explanation of the difference between the actual flowers and the order, along with a complimentary coupon.

[0143] If the comprehensive performance risk information indicates that the order falls under the category of Level 1 reshipment, and given that the flowers do not have substantial quality defects but only minor perceived differences, the system will trigger a low-cost reassurance strategy to avoid logistical waste caused by direct returns or reshipments. The system will automatically generate a push notification explaining the differences between the actual flowers and the order, and send it to the user via a client application (such as a mobile app or SMS). This explanation will objectively explain the reasons for the differences, such as the actual color possibly being slightly darker than the picture due to lighting conditions during transport, or the packaging being slightly compressed to protect the flowers. This aims to manage user expectations and reduce disappointment caused by information asymmetry. Simultaneously, a coupon of a certain amount will be issued to the user's account, offering a discount for future purchases to appease user sentiment and lock in repeat purchases, converting potential complaints into brand loyalty.

[0144] While implementing the above-mentioned reassurance strategies, the server sends delivery priority upgrade instructions or route adjustment instructions to the delivery platform, so that the delivery platform avoids road sections with high predicted bump intensity or long exposure time to high temperature; therefore, the first-level reassignment is adjusted within the delivery node without changing the supplier.

[0145] Step 5: If the reshipment is determined to be a Level 2 resend, implement a compensation strategy of providing the user with a partial refund and reshipping replacement flowers.

[0146] For an order determined as a secondary reallocation, since the physical structural integrity of the flower materials is damaged and cannot meet the initial consumption demand, the system executes a composite compensation strategy that combines partial refund and re-delivery of replacement flower materials. The partial refund mechanism provides immediate economic compensation to the user based on the degree of damage (for example, 30% to 50% of the order amount) to make up for psychological loss. As a remedial measure to fulfill the core obligations of the contract, re-delivering the replacement flower materials is not simply resending the same product of the original order. Especially for special categories such as spray-painted flowers, if the original delivery environment or process causes color fading or peculiar smell, re-delivering the same product may face repeated risks. Therefore, in some embodiments, the original spray-painted flowers can be replaced with ordinary fresh flowers with more stable quality and less prone to similar problems (for example, replacing blue enchantress roses with high-quality red roses) to ensure a good user experience. This strategy not only solves the current order fulfillment failure problem, but also effectively avoids the risk of repeated complaints caused by secondary delivery.

[0147] In specific implementation, the server selects an alternative fulfillment node according to the reallocation matching score of candidate merchants, sends an original fulfillment task termination instruction to the original merchant platform, sends an order acceptance instruction containing the standardized product fields of the original order to the alternative merchant platform, and sends a new pickup point, target delivery route and estimated delivery time to the delivery platform. After the delivery platform returns the new delivery task number, the server binds the new delivery task number with the original user platform order identifier, and updates the fulfillment node and estimated delivery time displayed on the user terminal, thereby completing the actual reallocation of the cross-platform order.

[0148] Step 6: When it is determined as tertiary reallocation, the reallocation strategy of full refund and original order cancellation is executed.

[0149] For the tertiary reallocation scenario where the flower materials lose ornamental value and have peculiar smell, which negatively affects the user experience, continuing to fulfill the order is meaningless and may cause serious customer complaints. The system immediately starts the highest-level emergency disposal strategy, executes the full refund process, and returns all the funds paid by the user in the original path to ensure fund security. At the same time, the original order is canceled, and all subsequent logistics delivery nodes are terminated to prevent the problematic products from continuing to circulate. In addition, a quality warning report is synchronously generated and fed back to the merchant platform or the quality management department, so as to investigate and rectify the supply chain links (such as processing batches, delivery routes or packaging types) that cause such serious abnormalities, and eliminate hidden dangers from the source.

[0150] Embodiment 11: In some embodiments, to illustrate the specific calculation process of comprehensive fulfillment risk classification and abnormal reallocation, h_1=0.40, h_2=0.60, and λ=0.20 are used as example calibration parameters. The parameters are only used to illustrate the calculation process, and are recalibrated according to the historical order data of the target flower material in practical applications.

[0151] Example 1: For a standard unpainted bouquet, the initial fulfillment deviation D_0 between the user and the merchant is 0.12, and the delivery risk R_d is 0.18. The calculated R = 0.40 × 0.12 + 0.60 × 0.18 + 0.20 × 0.12 × 0.18 = 0.16032, which is within the normal fulfillment range. The server maintains the original merchant node and the original delivery task.

[0152] Example 2: A bouquet of flowers has a D_0 of 0.28 and an R_d of 0.32, resulting in an R = 0.32192. It falls within the first-level reassignment range and does not meet the hard trigger condition. The server sends a priority upgrade and low-bump route adjustment instruction to the delivery platform and pushes a difference explanation to the user's device.

[0153] Example 3: A certain spray-painted bouquet has a D_0 of 0.48 and an R_d of 0.58, resulting in an R = 0.59568, which falls within the second-level reassignment range. Candidate Merchant A and Candidate Merchant B have reassignment matching scores of 0.7325 and 0.7950 respectively. The server selects Candidate Merchant B as the alternative fulfillment node, terminates the original delivery task, generates a new pickup task, and simultaneously executes a partial refund and replacement flower delivery according to the second-level reassignment strategy.

[0154] Example 4: For a sealed packaged flower bouquet with sprayed color, D_0 is 0.62 and R_d is 0.76, resulting in R=0.79824, which falls within the level 3 relocation range. Alternatively, even if R is below 0.75, if the odor mixing deviation O_s reaches 0.83, it is still classified as level 3 due to hitting the hard trigger condition. The server terminates the original fulfillment link, cancels the original order, and executes a full refund, while simultaneously recording the abnormal processing batch, packaging type, and delivery route in the quality warning record.

[0155] As can be seen from the above examples, the risk range is used to handle comprehensive anomalies formed by the accumulation of multiple factors, the hard trigger condition is used to identify a single serious defect, and the candidate merchant matching score is used to select the node that can complete the alternative delivery within the remaining fulfillment time during the secondary re-delivery, thus forming a closed loop from status quantification, risk classification to cross-platform instruction execution.

[0156] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for aligning fulfillment status and resending abnormal flower orders across platforms, characterized in that, include: Obtain the user platform order information set and the merchant platform product information set. Based on the user platform order information set and the merchant platform product information set, analyze the perceptual deviation characteristics of floral materials caused by human processing and natural evolution, as well as the changes of floral materials under physical combination and structural constraints, to obtain the user-merchant fulfillment deviation information set. Obtain the delivery platform status information set, and based on the delivery platform status information set and the user-merchant fulfillment deviation information set, analyze the ability of flowers to maintain their quality during delivery and the characteristics of quality re-change caused by environmental factors to obtain the delivery risk assessment information set. Integrate the user-merchant fulfillment deviation information set and the delivery risk assessment information set, classify abnormal reassignment levels, and output and execute cross-platform flower order abnormal reassignment strategies.

2. The method according to claim 1, characterized in that, The process of constructing the user-merchant performance deviation information set includes: The user platform order information set includes the type of bouquet and the quantity of flowers. The merchant platform's product information set includes information on flower material processing and flower material warehousing. Based on the type of bouquet, the quantity of flowers, and the processing information of the flowers, the effects of packaging methods on the size and shape of the flowers, as well as the effects of spraying on the color and smell of the flowers, are analyzed to obtain the human-induced size and shape deviations and the spraying color and smell deviations. Based on the quantity of flowers, the type of bouquet, and the flower storage information, the effects of storage temperature and humidity on the freshness and natural color of flowers are analyzed to obtain the natural evolution color deviation. By integrating the human-caused size and shape deviations, the color and odor deviations of the sprayed paint, and the color deviations of natural evolution, a user-merchant performance deviation information set is constructed.

3. The method according to claim 2, characterized in that, The process of constructing the artificial size and shape deviation includes: Based on the quantity of flowers and the processing information of the flowers, the physical compression characteristics of the external binding and wrapping methods on the original extended state of the flowers, as well as the physical restrictions of the internal space of the packaging container on the spreading angle of the flowers, are analyzed to obtain information on compression distortion. Based on the types of bouquets and the information on compression distortion, the differential tolerance characteristics of stem flexibility and flower fragility under packaging compression of different flowers are analyzed to obtain compression tolerance information. By integrating the pressure distortion information and the compression tolerance information, the artificial dimensional and morphological deviation is constructed.

4. The method according to claim 2, characterized in that, The process of constructing the color and odor deviation of the sprayed paint includes: Based on the types of bouquets and the processing information of the flowers, the differences in the adhesion ability of the surface texture of the petals of different flowers to the sprayed pigments, as well as the masking characteristics of the original natural color of the flowers after the pigments are applied, are analyzed to obtain the color coverage deviation. Based on the quantity of the flowers and the processing information of the flowers, the uniform spraying characteristics of the sprayed pigment on the batch of flowers are analyzed, as well as the odor variation characteristics caused by the mixture of the residual odor after the solvent in the pigment evaporates and the fragrance of the flowers themselves, to obtain the odor mixing deviation. The color coverage deviation and the odor mixing deviation are integrated to construct the spray color odor deviation.

5. The method according to claim 2, characterized in that, The process of constructing the naturally evolved color deviation includes: Based on the quantity of flowers and the storage information of flowers, the oxidation and discoloration characteristics of the flower cuts and epidermis due to contact with air are analyzed to obtain oxidation and discoloration information. Based on the oxidation discoloration information and the type of bouquet, the differences in the stability of petal pigments and the thickness of the epidermal wax layer of different floral materials are analyzed to determine the differential sensitivity characteristics of the color darkening and spot distribution caused by oxidation reaction, thus obtaining the natural evolution color deviation.

6. The method according to claim 5, characterized in that, The method involves analyzing the preservation capacity of flowers during delivery and the quality re-change characteristics caused by environmental factors, based on the delivery platform status information set and the user-merchant fulfillment deviation information set, to obtain a delivery risk assessment information set, including: The delivery platform status information set includes delivery route information, delivery time period, and delivery packaging type; Based on the delivery route information and the human-caused size and shape deviations, the further impact of transportation bumps on the size and shape of the flowers is analyzed to obtain transportation deformation information; Based on the transportation deformation information, combined with the delivery time period and the natural evolution color deviation, the influence of temperature changes during the delivery time period on the quality of the flowers is analyzed to obtain the quality change information during the time period; Based on the time period quality change information, combined with the delivery packaging type and the color and odor deviation of the sprayed paint, the influence of the packaging environment on the color and odor of the flowers is analyzed to obtain the delivery risk assessment information set.

7. The method according to claim 6, characterized in that, The process of constructing the transport deformation information includes: Based on the delivery route information, the vertical vibration frequency characteristics caused by differences in road surface smoothness in the transportation route, as well as the horizontal inertial characteristics during vehicle start-stop and turning processes, are analyzed to obtain bump and impact information. Based on the bump and impact information, combined with the human-induced size and shape deviation, the characteristics of flower head displacement caused by external impact on the basis of the original physical compression are analyzed to obtain the bouquet distortion information. Based on the bouquet distortion information, the mutual compression and friction characteristics between individual flower materials caused by displacement, as well as the curling and shedding trends of petal edges caused by collisions, are analyzed to obtain the transport deformation information.

8. The method according to claim 7, characterized in that, The process of constructing the time-period quality change information includes: Based on the transport deformation information, the damage to the epidermis of the compressed parts of the flower material and the water loss characteristics caused by the obstruction of the water supply channels inside the flower material are analyzed to obtain the flower material damage information. Based on the information on damaged flowers and the delivery time period, we analyzed the characteristics of changes in the respiration intensity of flowers caused by changes in ambient temperature during the delivery process, as well as the characteristics of accelerated water loss from damaged areas, to obtain physiological loss information. Based on the physiological depletion information and the natural evolutionary color deviation, the color deepening phenomenon of the damaged parts on the basis of the original oxidative discoloration and the color darkening phenomenon of the healthy parts due to water loss are analyzed to obtain the quality change information of the time period.

9. The method according to claim 8, characterized in that, Based on the time-period quality change information, combined with the delivery packaging type and the color and odor deviation of the spray paint, the influence of the packaging environment on the color and odor of the flowers is analyzed to obtain the delivery risk assessment information set, including: Based on the time period quality change information and the delivery packaging type, the limiting characteristics of the light transmittance and air permeability of the packaging material on the maintenance of the flower color, as well as the shielding or mixing effect of the packaging material's own adsorption or release properties on the flower odor, are analyzed to obtain interactive influence information. Based on the interaction information and the color and odor deviation of the sprayed pigment, the secondary mixing and crosstalk characteristics of the residual solvent volatiles of the sprayed pigment and the natural fragrance of the flowers in the packaging space are analyzed to obtain the mixing crosstalk information. The interaction impact information and the mixed crosstalk information are integrated to construct the delivery risk assessment information set.

10. The method according to claim 9, characterized in that, The process of integrating the user-merchant fulfillment deviation information set and the delivery risk assessment information set, classifying abnormal reassignment levels, and outputting and executing cross-platform flower order abnormal reassignment strategies includes: Based on the user-merchant fulfillment deviation information set and the delivery risk assessment information set, the expected quality degradation limit of the flowers when they arrive at the user is analyzed under the superposition of the transportation deformation information, the time period quality change information and the interaction impact information, so as to obtain comprehensive fulfillment risk information. Based on the comprehensive performance risk information, the corresponding abnormal reassignment level is determined, and the cross-platform flower order abnormal reassignment strategy is output and executed. The abnormal reassignment levels include Level 1 reassignment for flowers that have sensory differences but still meet basic delivery standards, Level 2 reassignment for flowers whose physical structural integrity is damaged, and Level 3 reassignment for flowers that have lost their ornamental value and have an odor. When the order is determined to be a Level 1 reassignment, a reassurance strategy is implemented by sending the user an explanation of the difference between the actual flowers and the order, along with a coupon. When the reassignment is determined to be a Level 2 reassignment, a compensation strategy of providing the user with a partial refund and re-delivering replacement flowers will be implemented; If the reassignment is determined to be at level three, a full refund and cancellation of the original order will be implemented.

Images