Feed storage dynamic decision method and system based on comprehensive warehouse system cooperation

Abstract

The present application relates to the technical field of feed storage management, in particular to a feed storage dynamic decision-making method and system based on comprehensive storage system coordination, the method comprising: obtaining storage property data of a feed storage center, feed supply time sequence data of a supplier and breeding cycle data, and preprocessing; predicting feed demand fluctuation based on the breeding cycle data, combining feed storage capacity constraints to generate an initial decision-making scheme including multiple feed plan storage time points; based on the feed supply time sequence data, dynamically adjusting the plan storage time points through compatibility verification and a preset coordination mechanism; constructing a decision-making model with the objective of minimizing comprehensive decision-making cost, and solving the optimal decision-making scheme of the feed storage center in combination with preset constraint conditions. The purpose is to realize dynamic decision-making management of multiple categories of feed in large-scale feed storage centers and improve the decision-making efficiency and accuracy of multiple categories of feed in large-scale feed storage centers.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of feed storage management, in particular to a feed storage dynamic decision-making method and system based on comprehensive storage system cooperation. BACKGROUND

[0002] In the field of feed storage management, large-scale storage centers, as the core nodes of the supply chain, usually also bear the core hub of connecting suppliers and breeding enterprises. With the development of livestock farming towards scale and intensification, the types of feed are highly subdivided, and different types of feed have significant differences in shelf life, temperature and humidity sensitivity, and physical properties, etc., which makes it difficult for traditional storage decision-making based on static rules to adapt to dynamic and diversified management needs.

[0003] The current industry generally uses storage decision-making methods that rely on intelligent decision-making of a single storage capacity indicator and artificial experience judgment to determine the storage location and quantity of various types of feed. Since this approach does not fully consider the actual needs of the storage center, feed suppliers, and breeding enterprises, such as real-time changes in storage capacity, changes in supply from suppliers, and changes in demand from breeders, the decision-making of the storage center, suppliers, and breeders is isolated and cannot be adjusted in coordination. This ultimately results in low decision-making efficiency and insufficient scheduling accuracy for large-scale feed storage centers, which hinders the dynamic management of multiple types of feed in large-scale feed storage centers. SUMMARY

[0004] In order to achieve dynamic decision-making management of multiple types of feed in large-scale feed storage centers and improve the efficiency and accuracy of storage decision-making for multiple types of feed in large-scale feed storage centers, the present application provides a feed storage dynamic decision-making method and system based on comprehensive storage system cooperation, which uses the following technical solutions:

[0005] The technical solution of the first aspect of the present application provides a feed storage dynamic decision-making method based on comprehensive storage system cooperation, which includes:

[0006] Obtain storage property data of the feed storage center, feed supply timing data of the supplier, and breeding cycle data, and preprocess them;

[0007] Based on the breeding cycle data, predict the fluctuation of feed demand, and generate an initial decision-making scheme including multiple planned storage times of feed based on the feed storage capacity constraint;

[0008] Based on the feed supply timing data, dynamically adjust the planned storage time through compatibility verification and a preset cooperation mechanism;

[0009] Construct a decision-making model with the goal of minimizing the comprehensive cost of decision-making, and solve the optimal decision-making scheme of the feed storage center based on the preset constraint conditions.

[0010] Further, the feed demand fluctuation is predicted based on the breeding cycle data, including:

[0011] Based on the pre-processed breeding cycle data, the actual demand amount time series of each type of feed in at least two complete animal husbandry cycles is extracted;

[0012] The periodic decay factor is adjusted according to the fluctuation intensity of the actual demand amount time series, and the baseline demand offset is set according to the global mean value of the actual demand amount time series;

[0013] The multiple periodic fluctuation components are assigned weight coefficients according to the historical influence proportion, and the fluctuation values of each component are calculated according to the adjusted periodic decay factor and the time decay;

[0014] The fluctuation components are superimposed and summed with the baseline offset to generate a demand prediction curve, and the feed prediction demand amount at each time point is extracted from the demand prediction curve according to the preset time granularity.

[0015] Further, an initial decision scheme including multiple planned storage time points of feed is generated in combination with the feed storage capacity constraint, including:

[0016] Based on the proportional relationship between the predicted demand amount and the total storage capacity, the decision cycle is divided into consecutive time windows;

[0017] The length of the time window is dynamically adjusted according to the shortest shelf life requirement of the feed category;

[0018] The planned storage time point set is generated according to the center time point of the time window.

[0019] Further, the compatibility verification includes:

[0020] The time point closest to the supplier's supply time point is located in the planned storage time point set;

[0021] The absolute time deviation between the supply time point and the located time point is calculated, and it is verified whether the current remaining capacity of the warehouse supports the supply amount;

[0022] If the verification is passed, the supply time point is replaced by the planned storage time point, and if the verification is not passed, a preset coordination mechanism is triggered.

[0023] Further, the preset coordination mechanism includes:

[0024] The supply amount interval of the feed storage center is dynamically calculated, including the lower limit value of the received amount and the upper limit value of the received amount;

[0025] The coordination request of the adjustable range of the time window and the supply amount interval is generated according to the supply amount interval, and is sent to the supplier;

[0026] Update supply timing data based on supplier adjustment responses.

[0027] Furthermore, based on the feed supply time series data, the planned storage time is dynamically adjusted through compatibility verification and a preset coordination mechanism, further comprising:

[0028] The storage priority is calculated based on the remaining shelf life, the demand urgency decay index and the critical state of the warehouse, and the storage time period and storage warehouse resources of conflicting batches are dynamically allocated according to the storage priority.

[0029] Furthermore, a decision-making model is constructed with the goal of minimizing the comprehensive cost of decision-making, including:

[0030] Define three-dimensional decision variables, including feed type, storage time, and storage bin allocation relationship;

[0031] Calculate storage delay cost based on delay duration and unit time cost coefficient;

[0032] The supply deviation cost is calculated as a linear function of the absolute deviation between the storage moment and the supply moment;

[0033] Feed loss costs were calculated based on the temperature response function of the Arrhenius equation.

[0034] Furthermore, the mathematical expression of the objective function of the decision model with the goal of minimizing the comprehensive cost of decision-making is:

[0035]

[0036] Where, express Ton categories are Is the feed in the first target storage time is allocated to the storage bin ; represents the storage delay cost coefficient; represents the supply deviation cost coefficient; represents the feed loss cost coefficient; Indicates the original planned storage time of the feed; Indicates feed Supplier supply time; Indicates the temperature Feed loss rate.

[0037] Furthermore, the preset constraints include:

[0038] Storage silo capacity constraint, configured so that the total amount of feed stored in a single silo does not exceed its capacity;

[0039] Time mutual exclusion constraint, configured so that a single storage bin can only store one type of feed at the same target storage time;

[0040] compatibility constraint configured to feed can only be assigned to silos supporting its type;

[0041] demand satisfaction constraint configured to total amount of each type of feed assignment equals to predicted demand amount;

[0042] supply window constraint configured to feed storage time is not earlier than supply start time.

[0043] The technical scheme of the second aspect of the present application provides a feed storage dynamic decision system based on comprehensive warehouse system coordination, which adopts the feed storage dynamic decision method based on comprehensive warehouse system coordination of the first aspect of the present application. The system comprises:

[0044] a data acquisition and preprocessing module configured to acquire silo attribute data of a feed storage center, feed supply time sequence data of a supplier and breeding cycle data, and to preprocess the data;

[0045] an initial decision module configured to predict feed demand fluctuation based on the breeding cycle data, and to generate an initial decision scheme comprising multiple feed plan storage times in combination with feed storage capacity constraints;

[0046] a feed storage time adjustment module configured to dynamically adjust the plan storage times based on feed supply time sequence data through compatibility verification and a preset coordination mechanism;

[0047] a decision module configured to construct a decision model with the objective of minimizing comprehensive cost, and to solve an optimal decision scheme of the feed storage center in combination with preset constraint conditions.

[0048] The present application has the following beneficial effects:

[0049] The feed storage dynamic decision method based on comprehensive warehouse system coordination provided by the present application acquires silo attributes, supply time sequence and breeding cycle data and preprocesses the data, predicts demand fluctuation based on the breeding cycle to generate an initial plan storage time scheme in combination with storage capacity, dynamically adjusts the plan storage times according to supply time sequence data through compatibility verification and a coordination mechanism, and finally constructs a decision model with the objective of minimizing comprehensive cost and solves an optimal scheme in combination with constraint conditions. The present application realizes three-party collaborative decision of the warehouse center, the supplier and the breeding enterprise, coordinates multiple types of feed storage time and location assignment, improves warehouse decision efficiency, optimizes feed resource utilization rate by minimizing decision cost, and matches supply and storage capacity. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, below will briefly introduce the drawings needed in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

[0051] Figure 1 The method flowchart of the feed storage dynamic decision method based on the comprehensive warehouse system cooperation provided by an embodiment of the present application;

[0052] Figure 2 The structural schematic diagram of the feed storage dynamic decision system based on the comprehensive warehouse system cooperation provided by an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined application purposes, below, the specific embodiments, structures, features and effects of the feed storage dynamic decision method and system based on the comprehensive warehouse system cooperation according to the present application are described in detail in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" does not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0055] Below, the specific scheme of the feed storage dynamic decision method and system based on the comprehensive warehouse system cooperation provided by the present application is specifically described in combination with the drawings.

[0056] The present application takes a large feed storage center as an example, the feed storage center is configured with a plurality of feed storage bins for storing different types of feed; the target is to realize the dynamic decision management of the large feed storage center multi-type feed and the collaborative decision among the large feed storage center, the supplier and the breeding enterprise; to improve the feed storage decision efficiency and accuracy of the large feed storage center; finally, the feed plan storage time, the supplier supply time, the feed demand, the storage capacity and other data are input into the decision optimization model, and the above-mentioned preset constraint conditions are combined to solve through a mixed integer programming solver and other tools, and the dynamic decision result of the large feed storage center multi-type feed under the condition of meeting the actual situation and the lowest comprehensive storage decision cost is obtained.

[0057] Please refer to Figure 1, which shows a method flow chart of a feed storage dynamic decision-making method based on integrated storage system collaboration provided by one embodiment of the present invention, the method comprising:

[0058] Step S100: Obtaining the storage attribute data of the feed storage center, the supplier's feed supply time series data and breeding cycle data, and performing pre-processing;

[0059] Step S100 specifically includes:

[0060] Step S110: Obtain the storage-related data, feed supply data and feed breeding data of the feed storage center; define the storage warehouse set as , The total number of storage bins; storage bin related data includes storage bin capacity , indicating storage The maximum capacity, in tons; storage bin coordinates , indicating storage Plane coordinate position, storage bin compatible feed type vector , Indicates storage warehouse The storage category is of feed, It is a collection of feed categories; it also collects temperature data of each storage bin; feed supply data includes feed supply time , indicating the feed provided by the supplier Arrival time of the feed storage center and the amount of feed provided by the supplier Breeding data includes livestock cycle data of breeding enterprises, which is used to predict the fluctuation of feed demand of breeding enterprises;

[0061] Step S120: Perform spatiotemporal alignment preprocessing on the storage silo-related data, feed supply data, and feed breeding data; the spatiotemporal alignment preprocessing includes unified spatiotemporal processing of the storage silo-related data, feed supply data, and breeding data; for the time dimension, calibrate the timestamps of data from different sources to a unified time coordinate system to ensure the consistency of the supply time, breeding cycle time point, and storage silo status time record; for the spatial dimension, match the storage silo coordinate data with the feed storage location requirements, and at the same time, combine the feed supply source and the geographical location of the breeding enterprise to construct a spatial association matrix.

[0062] Step S200: Predicting feed demand fluctuations based on the breeding cycle data, and generating an initial decision plan including multiple planned feed storage times in combination with feed storage capacity constraints;

[0063] Step S200 specifically includes:

[0064] Step S210: Based on the pre-processed breeding cycle data, extract the actual demand time series of each type of feed in at least two complete animal husbandry cycles; the actual demand time series can be expressed as , indicating feed The actual demand, Indicates a historical point in time;

[0065] Step S211: Adjust the periodic attenuation factor according to the fluctuation intensity of the actual demand time series, and set the baseline demand offset according to the global mean of the actual demand time series; specifically, calculate the demand time series Variance Characterizes the intensity of fluctuations, , the larger the variance, the faster it decays; the baseline demand offset can be expressed as:

[0066]

[0067] Where, is the baseline demand offset; is the total number of time points;

[0068] Step S212: Assign weight coefficients to multiple periodic fluctuation components according to their historical impact ratios, and calculate the fluctuation value of each component that decays over time based on the adjusted periodic attenuation factor; specifically, decompose the historical demand time series using the EM algorithm to obtain Gaussian components, the historical average proportion of each component is its contribution coefficient to the category feed demand , reflecting the influence weight of the periodic fluctuation component in the historical demand; for each Gaussian component, its central moment Corresponding to key time points in the livestock cycle, such as livestock growth stage transition points, using adjusted attenuation factors , calculate the fluctuation value of this component, and the predicted demand can be finally expressed as:

[0069]

[0070] Where, express Feed at the moment Forecast demand; Indicates the Gaussian components for categories Contribution coefficient of feed requirement; Indicates the category The damping factor of feed demand fluctuations controls the width of the Gaussian function; represents the central moment of the kth Gaussian component, corresponding to the key time point in the livestock cycle; Indicates the number of Gaussian components, which can be determined according to the complexity of the livestock cycle;

[0071] Step S213: superimpose the fluctuation component and sum it with the reference offset to generate a demand prediction curve, and extract the predicted demand amount of each time point from the demand prediction curve according to the preset time granularity; in this embodiment, the actual demand amount time sequence of at least two complete livestock cycles is extracted, the decay factor is adjusted based on the fluctuation intensity, the reference offset is set based on the global mean value, the weight of the Gaussian component is distributed according to the historical influence proportion, and the fluctuation value is calculated, and finally the demand prediction curve is generated by superimposition, thereby realizing dynamic modeling of demand fluctuations of multiple types of feed. Through this method, the periodic change rule of demand in the breeding cycle can be accurately captured, and the initial planning storage time scheme is generated in combination with the warehouse capacity constraint, which not only considers the dynamic fluctuation characteristics of feed demand, but also meets the limitation condition of warehouse space, thereby ensuring that the initial decision scheme can fit the real demand fluctuation of the breeding enterprise and improving the forward-looking and rationality of the warehouse decision.

[0072] Step S220: based on the proportional relationship between the predicted demand amount and the total warehouse capacity, the decision cycle is divided into continuous time windows; based on the predicted demand amount and the warehouse capacity constraint, the length of each time window is determined to ensure that the total amount of feed stored in the window does not exceed the total capacity of the warehouse center, and the warehouse capacity constraint can be represented as:

[0073]

[0074] This formula is used to constrain the window , and the total amount of feed demand is less than or equal to the total effective warehouse capacity; represents the maximum capacity of the warehouse ;

[0075] Step S221: dynamically adjust the length of the time window according to the shortest shelf life requirement of the feed category; specifically, the shortest shelf life of the feed category can be represented as , that is, the shortest shelf life of the feed ; if , the window is contracted;

[0076] Step S222: generate a set of planned storage time points according to the center time point of the time window; for each time window, the center time point is selected as the planned storage time point to form a set of planned storage time points, which can be represented as: , is the number of planned storage time points;

[0077] In this embodiment, the length of the time window Ensure that the total amount of demand points in the window does not exceed the effective storage capacity, and shrink the window according to the shortest shelf life constraint to avoid feed expiration, and finally construct a plan storage time set at the center time of the window, which realizes the feasible initial decision scheme under the dual constraints of storage capacity and shelf life.

[0078] Step S300: Based on the feed supply timing data, the compatibility check and the preset coordination mechanism are used to dynamically adjust the planned storage time;

[0079] Step S300 specifically includes:

[0080] Step S310: Locate the time point closest to the supplier's supply time in the planned storage time set;

[0081] Step S320: Calculate the absolute time deviation between the supply time and the located time, and check whether the current storage remaining capacity supports the supply amount; calculate the time difference between the supplier's supply time and the planned storage time , find the planned storage time with the smallest time difference, which can be expressed as:

[0082]

[0083] In the formula, represents the planned storage time with the smallest absolute time difference;

[0084] If the time difference , is the time window threshold, and the value is , it is considered that the supply time is within the time window of the planned storage time; the storage capacity check includes calculating the available capacity of each storage at the supply time; if the supply time is within the time window and the storage capacity allows, directly use the supply time as the target storage time; otherwise, trigger the coordination mechanism to feedback the adjustment request to the supplier, including: supply time adjustment range: ; supply amount adjustment interval: , ;

[0085] Step S330: If the check passes, replace the supply time with the planned storage time, if the check fails, trigger the preset coordination mechanism; wherein the preset coordination mechanism includes: dynamically calculating the supply amount interval of the feed storage center, including the lower limit of the received amount and the upper limit of the received amount; generating a time window adjustable range ​and the supply interval, and sends to the supplier; then updates the supply timing data according to the adjustment response of the supplier; if the supplier agrees to adjust, updates the supply plan, and then returns to step S310 for re-verification; if the supplier cannot adjust, initiates emergency procurement, and processes the new supplier's supply time into the data set in step S100 for re-processing.

[0086] When the warehouse is overloaded due to multiple batches, step S300 further includes:

[0087] Step S340: Calculate the storage priority based on the remaining shelf life proportion, demand urgency decay index, and warehouse critical state, and dynamically allocate the storage time period and storage resource of the conflict batch according to the storage priority; in this embodiment, if the supply time is concentrated, the storage time period is dynamically allocated according to the feed priority; the priority calculation can be represented as:

[0088]

[0089] In the formula, represents the storage priority of the feed of category . , , is a weight coefficient, corresponding to the importance of shelf life, demand urgency, and warehouse critical state, respectively; represents the shelf life of the feed of category . represents the current storage time; represents the effective use period of the feed of category , which represents the longest allowed storage time from production to the beginning of deterioration; represents the warehouse critical state indicator function, which takes the value of 1 when the warehouse capacity is close to the upper limit, otherwise 0; is the shelf life remaining ratio, the larger the value, the higher the priority; is the demand urgency, the longer the storage time, the urgency index decays; is the warehouse capacity critical state, which forces to increase the priority when full; according to in descending order, the storage time period and storage resource are allocated to high-priority feed first to ensure that feed with short shelf life, high demand urgency, or critical warehouse capacity is stored first;

[0090] In summary, the embodiment realizes quick adaptation of supply time through double verification of time window threshold and storage capacity, triggers a collaborative mechanism to dynamically generate an acceptable supply amount interval and time adjustment window for those who fail the verification, and flexibly updates the plan in combination with the supplier's response; for multi-batch conflicts, resources are allocated based on a priority model, improving the supply deviation absorption capacity and warehouse overload emergency efficiency, ensuring that high-priority feed is stored, solving the problem of resource allocation when there are multi-batch conflicts, and achieving dynamic robustness optimization of three-party collaboration.

[0091] Step S400: Construct a decision model with the objective of minimizing the comprehensive cost of decision-making, and solve the optimal decision scheme of the feed storage center combined with the preset constraint conditions;

[0092] Step S400 specifically includes:

[0093] Step S410: Define three-dimensional decision variables, including feed category, storage time, and storage allocation relationship; define three-dimensional decision variables , which represents whether the feed of category is allocated to storage at the target storage time; , is the adjusted planned storage time set; through this variable, the storage allocation relationship of each category of feed at different times can be accurately described, avoiding storage conflicts; the comprehensive cost in the embodiment includes: calculating the storage delay cost according to the delay length and the unit time cost coefficient; calculating the supply deviation cost according to the linear function of the absolute deviation between the storage time and the supply time; calculating the feed loss cost according to the temperature response function of the Arrhenius equation; the mathematical expression of the objective function of the decision model with the objective of minimizing the comprehensive cost of decision-making is:

[0094]

[0095] In the formula, represents whether the feed of category is allocated to storage at the target storage time; represents the storage delay cost coefficient, which is used to measure the unit cost of the feed storage time being later than the planned time; represents the supply deviation cost coefficient, which is used to measure the unit cost of the deviation between the target storage time and the supplier's supply time; represents the feed loss cost coefficient, which is used to measure the unit cost of the characteristic loss during the storage of the feed; represents the original planned storage time of the feed, which only generates cost when the actual storage time is later than ; Indicates feed Supplier supply time; Indicates the temperature The feed loss rate under these conditions can be calculated based on the Arrhenius equation;

[0096] The preset constraints include:

[0097] The silo capacity constraint is configured so that the total amount of feed stored in a single silo does not exceed its capacity, which can be expressed as:

[0098]

[0099] Where, Represents any storage bin;

[0100] The time mutual exclusion constraint is configured so that a single storage bin can only store one type of feed at the same target storage time, which can be expressed as:

[0101]

[0102] A compatibility constraint, which configures feed to be allocated only to silos that support its type, can be expressed as:

[0103]

[0104] Where, Indicates storage warehouse Does it support feed? ;

[0105] The demand satisfies the constraint and the total amount of feed allocated to each category is equal to the predicted demand, which can be expressed as:

[0106]

[0107] The supply window constraint is configured so that the feed storage time is no earlier than the supply start time, which can be expressed as:

[0108]

[0109] Where, The start time of the supply window promised by the supplier; based on the above constraints, the above integer programming model is solved using a mixed integer programming solver, and the optimal solution is obtained by searching for a combination of decision variable values that satisfy all the constraints using an optimization algorithm to minimize the objective function value; the optimal decision scheme is presented in the form of a five-tuple set, including the feed category, the originally planned time, the adjusted time, the allocated silo, and the adjustment reason, while also providing details of the supply time deviation, the delay cost and the loss cost of each feed category, as well as a silo utilization heat map, to provide a visual decision basis for warehouse management. The present embodiment accurately depicts the allocation relationship between feed, time, and silo by defining three-dimensional decision variables, constructs a comprehensive optimization objective function including storage delay cost, supply deviation cost, and feed loss cost, and obtains the optimal decision scheme using a mixed integer programming solver in combination with multiple constraints such as silo capacity and time exclusion, thereby achieving cost optimization and accurate scheduling of resources in feed warehouse decision-making. The objective function quantifies the delay loss, the time deviation of supply and storage, and the impact of temperature on feed loss; this step converts warehouse decision-making into a quantifiable optimization problem through mathematical modeling, and the final output of the five-tuple scheme decision result not only minimizes the comprehensive cost of decision-making, but also ensures efficient use of silo resources, while providing data support for the three-way collaboration of warehouse centers, suppliers, and breeding enterprises, significantly improving the scientificity and accuracy of multi-category feed warehouse decision-making.

[0110] In summary, the feed warehouse dynamic decision-making method based on the collaboration of the comprehensive warehouse system can predict feed demand fluctuations based on breeding cycle data and generate an initial decision scheme in combination with warehouse capacity, which can anticipate the needs of breeding enterprises and warehouse capacity, and avoid storage chaos caused by unclear demand. Further, through compatibility verification and a preset collaboration mechanism, the planned storage time is dynamically adjusted, realizing the two-way interaction between the warehouse center and the supplier, effectively solving the deviation problem between the supply time and the planned storage time, and enhancing the adaptability of the warehouse system to dynamic supply. Finally, by constructing a decision-making model with the objective of minimizing the comprehensive cost of decision-making and combining the preset constraints to solve the optimal scheme, the allocation and utilization of feed resources are optimized from the cost perspective, while various constraints ensure the feasibility and rationality of the decision scheme. This method realizes the collaborative decision-making of the warehouse center, the supplier, and the breeding enterprise, and in the process of coordinating the storage time and location allocation of multi-category feed, not only improves the efficiency of warehouse decision-making, but also matches the supply and warehouse capacity through cost optimization and rational resource allocation, meeting the needs of dynamic management of multi-category feed in large feed warehouses.

[0111] Please refer to Figure 2 which shows the structure of the feed warehouse dynamic decision-making system based on the collaboration of the comprehensive warehouse system according to an embodiment of the present application, which comprises:

[0112] The data acquisition and preprocessing module is configured to acquire silo attribute data of the feed storage center, feed supply time sequence data of the supplier and breeding cycle data, and perform preprocessing;

[0113] The initial decision module is configured to predict feed demand fluctuation based on the breeding cycle data, and generate an initial decision scheme including a plurality of feed storage time points in combination with feed storage capacity constraints;

[0114] The feed storage time point adjustment module is configured to dynamically adjust the planned storage time point based on the feed supply time sequence data through compatibility verification and a preset coordination mechanism.

[0115] The decision module is configured to construct a decision model with the objective of minimizing comprehensive decision cost, and solve the optimal decision scheme of the feed storage center in combination with preset constraint conditions.

[0116] It should be noted that the above-mentioned embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0117] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments.

Claims

1. A dynamic decision-making method for feed storage based on the coordination of an integrated storage system, characterized in that: The method comprises: Obtain the storage attribute data of the feed storage center, the supplier's feed supply time series data and breeding cycle data, and perform preprocessing; Predicting feed demand fluctuations based on the breeding cycle data and generating an initial decision plan including multiple planned feed storage times in combination with feed storage capacity constraints; Based on the feed supply time series data, the planned storage time is dynamically adjusted through compatibility verification and a preset coordination mechanism; Construct a decision-making model with the goal of minimizing the comprehensive decision-making cost, and solve the optimal decision-making plan for the feed storage center based on the preset constraints; The predicting of feed demand fluctuations based on the breeding cycle data includes: Based on the pre-processed breeding cycle data, the actual demand time series of various feeds in at least two complete animal husbandry cycles are extracted; the actual demand time series is expressed as , indicating feed The actual demand, Indicates a historical point in time; Adjust the periodic attenuation factor according to the fluctuation intensity of the actual demand time series, set the benchmark demand offset according to the global mean of the actual demand time series, and calculate the demand time series. Variance Characterizes the intensity of fluctuations, , Indicates the category The attenuation factor of feed demand fluctuations, the baseline demand offset is expressed as: Where, is the baseline demand offset; is the total number of time points; Assign weight coefficients to multiple periodic fluctuation components according to their historical impact ratios, and calculate the fluctuation value of each component that decays over time based on the adjusted periodic attenuation factor. Decompose the historical demand time series using the EM algorithm to obtain Gaussian components, the historical average proportion of each component is its contribution coefficient to the category feed demand , reflecting the influence weight of the periodic fluctuation component in the historical demand; for each Gaussian component, its central moment Corresponding to key time points in the livestock cycle, using adjusted decay factors , calculate the fluctuation value of this component, and the predicted demand is finally expressed as: Where, express Feed at the moment Forecast demand; Indicates the Gaussian components for categories Contribution coefficient of feed requirement; Indicates the category Attenuation factor for fluctuations in feed demand; represents the central moment of the kth Gaussian component; represents the number of Gaussian components; Indicates the baseline demand offset; The fluctuation components are superimposed and summed with the baseline offset to generate a demand forecast curve, and the feed demand forecast at each time point is extracted from the demand forecast curve according to the preset time granularity; The generating of an initial decision plan including a plurality of planned feed storage times in combination with the feed storage capacity constraint includes: Based on the relationship between the predicted demand and the total storage capacity, the decision cycle is divided into continuous time windows to ensure that the total amount of feed stored in the window does not exceed the total capacity of the storage center; Dynamically adjust the time window length based on the minimum shelf life requirements of the feed category; Generate a planned storage time set according to the central moment of the time window; The compatibility check includes: Locate the time point closest to the supplier's supply time in the planned storage time set; Calculate the absolute time deviation between the supply time and the positioning time, and verify whether the current storage remaining capacity supports the supply; If the verification passes, the supply time will be replaced with the planned storage time. If the verification fails, the preset coordination mechanism will be triggered.

2. The feed storage dynamic decision-making method based on integrated storage system collaboration according to claim 1, characterized in that: Preset coordination mechanisms, including: Dynamically calculate the supply range of the feed storage center, including the lower limit and upper limit of the receiving amount; Generate a collaborative request for the adjustable time window range and the supply range based on the supply range, and send it to the supplier; Update supply timing data based on supplier adjustment responses.

3. The feed storage dynamic decision-making method based on integrated storage system collaboration according to claim 2, characterized in that: Based on the feed supply time series data, the planned storage time is dynamically adjusted through compatibility verification and a preset coordination mechanism, further comprising: The storage priority is calculated based on the remaining shelf life ratio, the demand urgency decay index, and the storage criticality. The storage time period and storage warehouse resources of conflicting batches are dynamically allocated according to the storage priority. The priority calculation is expressed as: Where, Indicates the category Storage priority of feed; 、 、 are weight coefficients, corresponding to the importance of shelf life, demand urgency, and storage criticality respectively; Indicates the category The shelf life of the feed; Indicates the current stored time; Category is The effective use period of the feed is the maximum permissible storage time from production to the beginning of deterioration; It represents the storage critical state indicator function, which takes the value of 1 when the storage capacity is close to the upper limit, otherwise it takes the value of 0; The remaining ratio of the shelf life, the larger the value, the higher the priority; The urgency of demand. The longer the storage time, the urgency exponentially decreases. The storage capacity is critical and the priority is forcibly increased when it is fully loaded.

4. The feed storage dynamic decision-making method based on integrated storage system collaboration according to any one of claims 1 to 3, characterized in that: Construct a decision-making model with the goal of minimizing the overall cost of decision-making, including: Define three-dimensional decision variables, including feed category, storage time and storage warehouse allocation relationship; define three-dimensional decision variables ,express Ton categories are Is the feed in the first target storage time is allocated to the storage bin ; , Store the set of moments for the adjusted plan; Calculate storage delay cost based on delay duration and unit time cost coefficient; The supply deviation cost is calculated as a linear function of the absolute deviation between the storage moment and the supply moment; Feed loss costs were calculated based on the temperature response function of the Arrhenius equation.

5. The feed storage dynamic decision-making method based on integrated storage system collaboration according to claim 4, characterized in that: The mathematical expression of the objective function of the decision-making model with the goal of minimizing the comprehensive cost of decision-making is: Where, express Ton categories are Is the feed in the first target storage time is allocated to the storage bin ; represents the storage delay cost coefficient; represents the supply deviation cost coefficient; represents the feed loss cost coefficient; Indicates the original planned storage time of the feed; Indicates feed Supplier supply time; Indicates the temperature Feed loss rate.

6. The feed storage dynamic decision-making method based on integrated storage system collaboration according to claim 5, characterized in that: The preset constraints include: Storage silo capacity constraint, configured so that the total amount of feed stored in a single silo does not exceed its capacity; Time mutual exclusion constraint, configured so that a single storage bin can only store one type of feed at the same target storage time; Compatibility constraints, configured so that feed can only be distributed to silos that support its type; The demand satisfies the constraint, and the total amount of feed allocated to each category is configured to be equal to the predicted demand; The supply window constraint is configured so that the feed storage time is no earlier than the supply start time.

7. A feed storage dynamic decision-making system based on the coordination of an integrated storage system, characterized in that: A method for implementing a dynamic feed storage decision-making method based on integrated storage system collaboration according to any one of claims 1 to 6, the system comprising: A data acquisition and preprocessing module is configured to acquire storage attribute data of the feed storage center, supplier feed supply time series data and breeding cycle data, and perform preprocessing; an initial decision module configured to predict feed demand fluctuations based on the breeding cycle data and generate an initial decision plan including multiple feed storage time plans in combination with feed storage capacity constraints; A feed storage time adjustment module is configured to dynamically adjust the planned storage time based on feed supply time series data through compatibility verification and a preset coordination mechanism; The decision-making module is configured to build a decision-making model with the goal of minimizing the comprehensive decision-making cost, and solve the optimal decision-making plan for the feed storage center in combination with preset constraints.

Images