Logistics distribution time window dynamic allocation method, apparatus and device, and storage medium

Abstract

The invention relates to the technical field of logistics distribution, and discloses a logistics distribution time window dynamic allocation method and device, equipment and a storage medium. The method comprises the steps of obtaining multi-source distribution data of a to-be-distributed area, and performing quartile bit distance anomaly detection and time sequence normalization to obtain standard multi-source distribution data; performing multi-period order feature dimension association and order priority calculation on the standard multi-source delivery data to obtain an order service level; performing region density calculation and distribution task similarity matching on the order service level to obtain a distribution task difficulty matrix; performing time sequence decomposition calculation and path planning on the distribution task difficulty matrix to obtain an order distribution time interval; and performing conflict detection and secondary time adjustment on the order delivery time interval to obtain a final delivery time allocation scheme. According to the invention, through multi-dimensional data processing and multi-level time optimization, dynamic and accurate distribution of the distribution time window is realized, the logistics distribution efficiency is improved, and the distribution cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of logistics distribution technology, and in particular to a logistics distribution time window dynamic allocation method, device, equipment and storage medium. Background Art

[0002] In the field of logistics and distribution technology, the reasonable allocation of time windows is the key to improving distribution efficiency and customer satisfaction, while order timeliness and resource scheduling are the most challenging stages in time window allocation. Accurate allocation of delivery time windows is crucial for logistics companies, distribution centers, and operations management, as it directly affects distribution cost control, service quality, and overall operational efficiency. Therefore, an effective time window dynamic allocation method to balance the needs of multiple parties is essential to ensure the efficient operation and service level of the logistics distribution system.

[0003] Nowadays, data mining techniques are used to analyze historical order data, establish distribution models to evaluate resource requirements, or try to apply artificial intelligence technology to the time window allocation process to better predict order distribution and optimize delivery routes. However, these methods still face challenges in integrating multi-source distribution data, processing dynamic order characteristics, and adapting to regional differences. In addition, these allocation methods often ignore some unique characteristics of logistics distribution, such as order timeliness, regional distribution pressure, road network accessibility, etc., which may have a significant impact on the reasonable allocation of time windows. That is, the existing logistics distribution time window allocation method ignores the dynamic balance of multi-dimensional distribution characteristics and regional distribution resources, so that the final time window allocation scheme is difficult to meet the actual distribution needs. Summary of the invention

[0004] The main purpose of the present invention is to solve the problem that the existing logistics distribution time window allocation method ignores the multi-dimensional distribution characteristics and the dynamic balance of regional distribution resources, so that the final time window allocation scheme is difficult to meet the actual distribution needs.

[0005] The first aspect of the present invention provides a method for dynamically allocating a logistics distribution time window, and the method comprises: acquiring multi-source distribution data of a to-be-allocated area, and performing interquartile range anomaly detection and time series normalization on the multi-source distribution data to obtain standard multi-source distribution data; performing multi-period order feature dimension association on the standard multi-source distribution data to obtain order spatiotemporal distribution features, and performing order priority calculation on the order spatiotemporal distribution features to obtain order service levels; performing density calculation on the order service levels of the to-be-allocated areas to obtain regional distribution pressure values, and performing similarity matching of distribution tasks on the regional distribution pressure values ​​to obtain a distribution task difficulty matrix; performing distribution time series decomposition calculation on the distribution task difficulty matrix to obtain distribution demand forecast values, and performing distribution path planning calculation on the distribution demand forecast values ​​to obtain order distribution time intervals; performing distribution conflict detection on the order distribution time intervals to obtain a distribution time window initial plan, and performing secondary distribution time adjustment on the distribution time window initial plan to obtain a final distribution time allocation plan.

[0006] Optionally, in a first implementation method of the first aspect of the present invention, the multi-source distribution data is subjected to interquartile range anomaly detection and time series normalization to obtain standard multi-source distribution data, including: calculating the interquartile range of order delivery time for the multi-source distribution data to obtain a normal order time interval, and determining a set of abnormal orders with corresponding time period deviations in the normal order time interval; comparing the delivery time of the abnormal order set based on the historical order time of the area to be allocated to obtain a high-frequency delivery time period, and calculating the time deviation between the high-frequency delivery time period and each abnormal order in the abnormal order set to obtain a time adjustment interval; allocating delivery time period capacity for the corresponding delivery orders in the time adjustment interval to obtain the multi-source distribution data after the delivery order time is adjusted, and based on a preset maximum delivery threshold per unit time, performing multi-time period division of delivery time requirements and capacity constraint calculation on the multi-source distribution data after the delivery order time is adjusted to obtain standard multi-source distribution data.

[0007] Optionally, in a second implementation method of the first aspect of the present invention, the standard multi-source distribution data includes logistics order data, distribution resource status data, distribution road network data and distribution environment data, and the standard multi-source distribution data is associated with multi-time period order feature dimensions to obtain order spatiotemporal distribution characteristics, including: based on the area to be allocated, the logistics order data is grouped into distribution areas according to the time period order quantity to obtain regional order distribution, and based on the distribution road network data, the route distance of the regional order distribution is calculated to obtain a regional distribution radius; based on the distribution resource status data, the number of delivery personnel corresponding to the regional distribution radius is determined to obtain a regional capacity distribution status, and the regional capacity distribution status is divided into time periods to obtain a time period deliverable capacity value; based on the distribution environment data, the time period deliverable capacity value is matched with the distribution road network to obtain the regional access time, and the regional access time is aggregated for time periods to obtain the order spatiotemporal distribution characteristics.

[0008] Optionally, in a third implementation of the first aspect of the present invention, the order priority calculation of the order spatiotemporal distribution characteristics to obtain the order service level includes: determining the time period order density corresponding to the unit time order volume in the order spatiotemporal distribution characteristics, and performing ratio calculation on the time period order density based on the maximum delivery volume of the delivery vehicle in the delivery resource status data to obtain the time period load rate; performing weighted calculation on the time period load rate based on the time period deliverable capacity value to obtain the delivery load index, and accumulating the delivery load index in the delivery target area to obtain the regional time period pressure; calculating the delivery guarantee rate corresponding to the regional time period pressure based on a preset route distance coefficient and the delivery vehicle data in the delivery resource status data, and performing secondary delivery calibration on the delivery guarantee rate based on the delivery environment data to obtain the regional service capacity; dividing and weighted averaging the regional service capacity into multiple delivery levels based on a preset regional saturation index to obtain an initial order level, and performing level calibration on the initial order level based on the maximum delivery capacity corresponding to the area to be allocated to obtain the order service level.

[0009] Optionally, in a fourth implementation method of the first aspect of the present invention, the density calculation of the area to be assigned to the order service level to obtain a regional delivery pressure value includes: combining the order service level with the spatiotemporal distribution characteristics of the order to obtain a regional priority delivery sequence, and matching the regional priority delivery sequence with the regional delivery radius to obtain a delivery coverage range; calculating the ratio of the number of delivery orders within the delivery coverage range to the regional transportation capacity distribution status to obtain a regional task density, and combining the regional task density with the regional access time to obtain a regional delivery difficulty; associating the regional delivery difficulty with adjacent delivery sub-areas to obtain an inter-regional coordination coefficient, and summarizing the delivery resources for the inter-regional coordination coefficient to obtain a regional delivery pressure value.

[0010] Optionally, in a fifth implementation method of the first aspect of the present invention, the similarity matching of delivery tasks for the regional delivery pressure values ​​to obtain a delivery task difficulty matrix includes: adding the regional delivery pressure values ​​and the order volumes of adjacent regions in the regional order distribution to obtain the regional cluster order volumes, and based on the delivery resource status data, performing regional delivery resource comparison on the regional cluster order volumes to obtain cluster delivery capacity differences; sorting the cluster delivery capacity differences by regional access time to obtain regional delivery urgency, combining the regional delivery urgency with the regional access time to obtain inter-regional support feasibility; performing delivery grading on the inter-regional support feasibility and the regional delivery difficulty to obtain a delivery task dispatch sequence, and normalizing the delivery indicators of the delivery task dispatch sequence to obtain a delivery task difficulty matrix.

[0011] Optionally, in a sixth implementation method of the first aspect of the present invention, the delivery task difficulty matrix is ​​subjected to delivery time series decomposition calculation to obtain a delivery demand forecast value, including: dividing the delivery task difficulty matrix into multiple time periods to obtain a time period delivery demand table, and weighting the time period delivery demand table with the regional delivery pressure value to obtain a regional time-varying demand; based on the historical delivery data corresponding to the area to be assigned, the regional time-varying demand is compared with the delivery trend and associated with the delivery difficulty to obtain a delivery difficulty correction value, and the delivery difficulty correction value is matched with the regional transportation capacity status to obtain a resource demand forecast; the resource demand forecast is subjected to threshold constraints and a balance of the delivery status of each delivery sub-area to obtain a forecast benchmark for each area, and the forecast benchmark for each area is verified with the regional access time to obtain an actual deliverable quantity; based on the delivery resource status data, the regional delivery demand value corresponding to the actual deliverable quantity is calculated, and the regional delivery demand value is weighted to obtain a delivery demand forecast value.

[0012] The second aspect of the present invention provides a dynamic allocation device for logistics distribution time windows, and the dynamic allocation device for logistics distribution time windows includes: a data preprocessing module, which is used to obtain multi-source distribution data of a to-be-allocated area, and perform interquartile range anomaly detection and time series normalization on the multi-source distribution data to obtain standard multi-source distribution data; a feature extraction module, which is used to associate multi-period order feature dimensions on the standard multi-source distribution data to obtain order spatiotemporal distribution features, and perform order priority calculation on the order spatiotemporal distribution features to obtain order service levels; a task evaluation module, which is used to perform density calculation of the to-be-allocated area for the order service level to obtain regional distribution pressure values, and perform similarity matching of distribution tasks on the regional distribution pressure values ​​to obtain a distribution task difficulty matrix; a demand analysis module, which is used to perform distribution time series decomposition calculation on the distribution task difficulty matrix to obtain distribution demand forecast values, and perform distribution path planning calculation on the distribution demand forecast values ​​to obtain order distribution time intervals; a plan generation module, which is used to perform distribution conflict detection on the order distribution time interval to obtain a distribution time window initial plan, and perform secondary distribution time adjustment on the distribution time window initial plan to obtain a final distribution time allocation plan.

[0013] The third aspect of the present invention provides a logistics distribution time window dynamic allocation device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory to enable the logistics distribution time window dynamic allocation device to perform each step of the above-mentioned logistics distribution time window dynamic allocation method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium, in which instructions are stored, and when the computer-readable storage medium is run on a computer, the computer executes each step of the above-mentioned method for dynamically allocating logistics distribution time windows.

[0015] The above-mentioned method, device, equipment and storage medium for dynamic allocation of logistics distribution time windows. In the embodiment of the present invention, by acquiring multi-source distribution data of the area to be allocated, and performing interquartile range anomaly detection and time series normalization on the multi-source distribution data, standard multi-source distribution data is obtained; multi-period order feature dimension association is performed on the standard multi-source distribution data to obtain the order spatiotemporal distribution characteristics, and the order priority is calculated on the order spatiotemporal distribution characteristics to obtain the order service level; the density of the area to be allocated is calculated for the order service level to obtain the regional distribution pressure value, and the regional distribution pressure value is matched with the similarity of the distribution task to obtain the distribution task difficulty matrix; the distribution task difficulty matrix is ​​decomposed and calculated to obtain the distribution demand forecast value, and the distribution demand forecast value is calculated for distribution path planning to obtain the order distribution time interval; the order distribution time interval is detected for distribution conflict to obtain the initial plan of the distribution time window, and the initial plan of the distribution time window is adjusted for secondary distribution time to obtain the final distribution time allocation plan. Compared with the prior art, this application obtains standardized distribution data by multi-source collection and anomaly detection of logistics order data, distribution resource data, road network data and environmental data in the distribution area, and then extracts spatiotemporal features and calculates order priority for the acquired data to obtain the order service level; then obtains the distribution task difficulty matrix and performs time series decomposition through regional density calculation and task similarity matching; finally, path planning and conflict detection are performed based on distribution demand forecast, and the final time window allocation plan is output. Through hierarchical data processing and feature analysis, dynamic and accurate allocation of logistics distribution time windows is achieved, fully considering the dynamic characteristics and regional differences of the distribution system, and effectively improving the allocation efficiency; and adopting multi-level feature association and task evaluation strategies, it not only realizes the reasonable allocation of distribution resources, but also enhances the dynamic adjustment ability of the time window; in addition, through demand forecasting and solution optimization, the distribution efficiency and service quality are accurately balanced, thereby realizing the efficient dynamic allocation of logistics distribution time windows as a whole.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of a first embodiment of a method for dynamically allocating logistics distribution time windows in an embodiment of the present invention; Figure 2It is a schematic diagram of an embodiment of a device for dynamically allocating a time window for logistics distribution in an embodiment of the present invention; Figure 3 It is a schematic diagram of an embodiment of a device for dynamically allocating time windows for logistics distribution in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0021] To facilitate understanding of this embodiment, the specific process of the embodiment of the present invention is described below. Figure 1 The first embodiment of the method for dynamically allocating logistics distribution time windows in the embodiment of the present invention includes: 101. Obtain multi-source distribution data of the area to be distributed, and perform interquartile range anomaly detection and time series normalization on the multi-source distribution data to obtain standard multi-source distribution data; The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0022] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0023] In this embodiment, the interquartile range of the order delivery time is calculated for the multi-source delivery data to obtain the normal order time interval, and the abnormal order set with the corresponding time period deviation in the normal order time interval is determined; based on the historical order time of the area to be allocated, the delivery time of the abnormal order set is compared to obtain the high-frequency delivery time period, and the time deviation between the high-frequency delivery time period and each abnormal order in the abnormal order set is calculated to obtain the time adjustment interval; the delivery time period capacity is allocated to the corresponding delivery orders in the time adjustment interval to obtain the multi-source delivery data after the delivery order time is adjusted, and based on the preset unit time maximum delivery threshold, the multi-source delivery data after the delivery order time is adjusted is divided into multiple time periods for delivery time requirements and capacity constraint calculations are performed to obtain standard multi-source delivery data. The above-mentioned unit time maximum delivery threshold refers to the maximum number of orders that the delivery system can handle within a specific time unit (such as 30 minutes).

[0024] In practical applications, we first obtain the multi-source distribution data of the area to be allocated. These data include the original multi-source distribution data corresponding to the logistics order data, distribution resource status data, distribution road network data and distribution environment data. The logistics order data contains order number, product information, expected delivery time and other information, the distribution resource status data contains the real-time status information of the delivery vehicle and the delivery person, the distribution road network data contains the road network topology and real-time road condition information, and the distribution environment data contains weather conditions and special events and other information; then, we calculate the interquartile range of the order delivery time, and determine the upper and lower limits of the normal order time interval by calculating the first quartile Q1 and the third quartile Q3 of the order time, and combining the interquartile range IQR (IQR=Q3-Q1), so as to identify the abnormal order set outside the interval; then, we compare and analyze the abnormal orders with the historical order time data of the area to be allocated, and obtain the high-frequency delivery time period by counting the delivery frequency of historical orders in different time periods, and calculate the expected delivery time of each order in the abnormal order set. The time deviation between the high-frequency delivery time period and the delivery time period is determined, and a reasonable time adjustment interval is determined based on these deviation values ​​(wherein this interval must take into account both the timeliness requirements of the order and the rational use of delivery resources); after obtaining the time adjustment interval, the orders within the time adjustment interval are reallocated according to the delivery resource capacity of each time period. The allocation process takes into account constraints such as the number of delivery vehicles and the working hours of the delivery personnel to ensure that the order volume in each time period does not exceed the carrying capacity of the delivery resources, and the adjusted order data is integrated with other delivery data to form a new multi-source delivery data set; based on the pre-set maximum delivery threshold per unit time, the integrated multi-source delivery data is divided into multiple time periods, usually divided into 30 minutes as a basic time unit, and the order volume in each time period is calculated with capacity constraints to ensure that the order allocation in different time periods is within the carrying capacity of the delivery system, and finally standardized multi-source delivery data is obtained, which will serve as the basic data support for the subsequent dynamic allocation of delivery time windows. For example, a certain delivery area detected that the number of abnormal orders exceeded expectations during the 9:00-10:00 period. By analyzing historical data, it was found that 8:30-9:30 was a high-frequency delivery period in the area. The time deviation was calculated and some orders were adjusted to the 8:30-9:30 period. At the same time, it was ensured that the number of orders in each 30-minute period after adjustment did not exceed the delivery threshold of 20 orders. Based on the preset maximum delivery threshold per unit time, the adjusted multi-source delivery data was normalized and capacity constraints were calculated to obtain standardized multi-source delivery data.

[0025] 102. Correlate the multi-period order feature dimensions of the standard multi-source delivery data to obtain the order's spatiotemporal distribution features, and calculate the order priority based on the order's spatiotemporal distribution features to obtain the order service level; In this embodiment, based on the area to be allocated, the logistics order data is grouped into delivery areas according to the order quantity of time periods to obtain regional order distribution, and based on the distribution road network data, the route distance of the regional order distribution is calculated to obtain the regional distribution radius; based on the distribution resource status data, the number of delivery personnel corresponding to the regional distribution radius is determined to obtain the regional transportation capacity distribution status, and the regional transportation capacity distribution status is divided into time periods to obtain the delivery capacity value of the time period; based on the distribution environment data, the delivery capacity value of the time period is matched with the distribution road network to obtain the regional access time, and the regional access time is aggregated by time periods to obtain the order spatiotemporal distribution characteristics, and the standard multi-source distribution data includes logistics order data, distribution resource status data, distribution road network data and distribution environment data; determine the time period order quantity corresponding to the unit time order quantity in the order spatiotemporal distribution characteristics density, and based on the maximum delivery volume of the delivery vehicles in the delivery resource status data, the ratio calculation of the order density in the time period is performed to obtain the time period load rate; based on the deliverable capacity value of the time period, the load rate of the time period is weightedly calculated to obtain the delivery load index, and the delivery load index is accumulated in the delivery target area to obtain the regional time period pressure; based on the preset route distance coefficient and the delivery vehicle data in the delivery resource status data, the delivery guarantee rate corresponding to the regional time period pressure is calculated, and based on the delivery environment data, the delivery guarantee rate is secondary calibrated to obtain the regional service capacity; based on the preset regional saturation index, the regional service capacity is divided into multiple delivery levels and weighted averaged to obtain the initial order level, and based on the maximum delivery capacity corresponding to the area to be allocated, the initial order level is calibrated to obtain the order service level. The route distance coefficient mentioned above refers to the correction parameters that affect the actual delivery route length during the delivery process (such as: road curvature correction: the ratio of the actual road distance to the straight-line distance, reflecting the impact of the road curvature on the delivery distance, traffic detour correction: the correction value of the necessary detour distance caused by factors such as one-way streets and traffic control, road grade weight: the distance equivalent correction caused by the difference in driving efficiency of roads of different grades (such as highways, main roads, and secondary roads), etc.). The regional saturation index mentioned above refers to an important indicator for measuring the order processing capacity within the delivery area, which indicates the degree of utilization of delivery resources in the area.

[0026] In practical applications, the logistics order data is first grouped and counted according to the delivery area, and the orders are assigned to different delivery areas according to their delivery addresses. At the same time, the order volume in each area is counted at 30-minute time intervals to obtain the order distribution of each area in different time periods. Based on these regional order distribution data and combined with the distribution road network data, the actual route distance between the order delivery points in each distribution area is calculated, and the effective delivery radius of each area is determined by the order density and road network coverage (where this delivery radius must ensure both delivery efficiency and the rational use of distribution resources); then, based on the distribution resource status data, the number of available delivery personnel within the distribution radius of each area is counted. This statistical process takes into account the delivery personnel's work The working hours, rest time and other factors are taken into account to obtain the capacity distribution status of each area, and these capacity distribution status are divided according to time periods to calculate the actual deliverable capacity value in each time period. This calculation process will take into account constraints such as the delivery efficiency per unit time and the maximum delivery volume of the delivery personnel. Then, based on the distribution environment data, including real-time information such as weather conditions and traffic conditions, the capacity of the distribution road network in each area is evaluated, and the regional access time under different environmental conditions is calculated. The access time reflects the actual reachable time from the distribution center to each distribution point, and these access times are aggregated and analyzed according to time periods, comprehensively considering the order distribution, capacity status and environmental impact, and finally obtaining the order spatiotemporal distribution characteristics that can fully reflect the spatiotemporal characteristics of the delivery task.

[0027] Secondly, the order volume per unit time is extracted from the temporal and spatial distribution characteristics of the orders, and the order density of each time period is counted based on 30 minutes as the basic time unit. Based on the maximum delivery volume information of the delivery vehicles in the distribution resource status data, the ratio of the time period order density to the maximum delivery volume is calculated to obtain the time period load rate that reflects the actual delivery pressure of each time period, so as to reflect the degree of matching between the order volume and the delivery capacity. Then, combined with the delivery capacity value of the historical time period, the time period load rate is weightedly calculated, and the delivery difficulty coefficient of different time periods is considered in the calculation process, so as to obtain a more accurate delivery load index, and the delivery load index of each time period in the same distribution target area is accumulated to obtain the regional time period pressure value that reflects the delivery pressure of the entire area. This pressure value comprehensively considers the delivery pressure in both time and space dimensions. Then, based on the pre-set route distance coefficient, combined with the distribution resource status data, the time period load rate is weightedly calculated. According to the specific delivery vehicle information in the data, including the number of vehicles, load capacity and other parameters, the delivery guarantee rate under the regional time period pressure is calculated to reflect the possibility of completing the delivery task under the current distribution resource conditions, and according to the distribution environment data, including weather conditions, traffic congestion level and other real-time information, the delivery guarantee rate is recalibrated to obtain a regional service capacity value that is more in line with the actual situation; then based on the pre-set regional saturation index, the regional service capacity is divided into multiple levels, that is, through priority delivery, standard delivery and delayed delivery, and the initial order level is obtained by weighted average, and the final level calibration is carried out based on the maximum delivery capacity of the area to be allocated, and the actual carrying capacity and service quality requirements of the distribution resources are considered during the calibration to ensure that the final order service level can not only meet customer needs but also ensure the stable operation of the distribution system.

[0028] 103. Calculate the density of the area to be assigned for the order service level to obtain the regional distribution pressure value, and perform similarity matching of the distribution tasks on the regional distribution pressure value to obtain the distribution task difficulty matrix; In this embodiment, the order service level and the order spatiotemporal distribution characteristics are combined with the distribution area characteristics to obtain a regional priority distribution sequence, and the regional priority distribution sequence is matched with the regional distribution radius to obtain the distribution coverage; the number of distribution orders within the distribution coverage is ratio-calculated to the regional transportation capacity distribution state to obtain the regional task density, and the regional task density is combined with the regional access time to obtain the regional distribution difficulty; the regional distribution difficulty is associated with adjacent distribution sub-areas to obtain the inter-regional coordination coefficient, and the inter-regional coordination coefficient is used to summarize the distribution resources to obtain the regional distribution pressure. value; the regional distribution pressure value and the order volume of the adjacent area in the regional order distribution are accumulated to obtain the regional cluster order volume, and based on the distribution resource status data, the regional cluster order volume is compared with the regional distribution resources to obtain the cluster distribution capacity difference; the cluster distribution capacity difference is sorted by the regional access time to obtain the regional distribution urgency, and the regional distribution urgency is combined with the regional access time to obtain the feasibility of inter-regional support; the inter-regional support feasibility and the regional distribution difficulty are distributed and graded to obtain the distribution task dispatch sequence, and the distribution index of the distribution task dispatch sequence is normalized to obtain the distribution task difficulty matrix.

[0029] In practical applications, we first conduct a combined analysis of the order service level and the temporal and spatial distribution characteristics of the order. That is, by comprehensively considering the service level distribution and temporal and spatial distribution characteristics of the orders in each region, we form a sequence that reflects the regional distribution priority. This priority distribution sequence not only reflects the urgency of the order, but also reflects the concentration and distribution pattern of the orders in the region. We match this priority distribution sequence with the regional distribution radius calculated previously, and calculate the actual distribution coverage of each region. This coverage must ensure both distribution efficiency and service quality. We then count the actual number of delivery orders within the statistical range and calculate the ratio of it to the regional transportation capacity distribution status. This calculation process takes into account the number of delivery personnel, Factors such as working hours are taken into account to obtain the regional task density that reflects the intensity of regional distribution tasks, and this task density is combined with the previously calculated regional access time for analysis. By comprehensively considering the order density and actual delivery time, a more accurate regional distribution difficulty is obtained. This difficulty index fully reflects the complexity of regional distribution tasks; and then by calculating factors such as the difference in order volume and capacity distribution between adjacent regions, a coordination coefficient that reflects the coordination ability between regions is obtained (this coordination coefficient is of great significance for balancing distribution resources between regions and improving overall distribution efficiency), and based on this coordination coefficient, combined with the distribution resource situation of each region, a summary calculation is performed to finally obtain a value that accurately reflects the regional distribution pressure. For example, a region contains three sub-regions: distribution difficulty =3.88, =4.2, =3.5; correlation coefficient =0.3, =0.2, =0.1; Difficulty difference =0.32, =0.7, =0.38; σ=0.4, RC=80, RCmax=100, calculated using the regional distribution pressure value formula, the formula is: ; in, is the delivery difficulty of the ith sub-region, is the correlation coefficient between regions i and j, is the distribution difficulty difference between regions, σ is the resource pressure coefficient, RC is the current resource occupancy, and RCmax is the maximum resource capacity. The corresponding data is substituted into the formula to calculate the regional distribution pressure value: RP=(3.88×1.096+4.2×1.14+3.5×1.038)×exp(0.4×80 / 100)=13.47×1.37=18.45.

[0030] Secondly, the regional distribution pressure value and the order volume of the adjacent regions are accumulated and calculated, and the distribution pressure of each region and the orders of the surrounding regions are considered in the accumulation process, so as to obtain the regional cluster order volume that reflects the overall distribution demand of the regional group. For example, the logistics order volume of a central area is 100, the pressure value is 0.8, the weight is 1.2, the order volumes of the two adjacent regions are 80 and 60 respectively, the impact factors are 0.6 and 0.5, and the distance attenuation coefficients are 0.9 and 0.8. The weighted accumulation formula is used: ; Among them, CVO is the regional cluster order volume, is the central area weight coefficient, is the regional distribution pressure value, is the regional order volume, is the adjacent region impact factor, is the order volume in the adjacent area, is the distance attenuation coefficient between regions, and the regional cluster order volume is calculated using the weighted accumulation formula: 1.2×0.8×100+(0.6×80×0.9+0.5×60×0.8)=139.2; then based on the distribution resource status data, including the number of available distribution vehicles, delivery personnel working hours and other information, the regional cluster order volume is compared with the distribution resource capacity, and the difference in cluster distribution capacity is calculated to reflect the distribution resource supply and demand balance of the regional group; then, combined with the previously calculated regional access time, it is sorted. This sorting process will take into account the actual travel time between different regions, so as to obtain the regional distribution urgency that reflects the urgency of the distribution tasks in each region, and this urgency is combined with the regional access time for analysis. That is, by comprehensively considering the time urgency and actual accessibility, possible inter-regional support plans are calculated to obtain the inter-regional support feasibility, which reflects the possibility of distribution resource allocation between different regions; then, based on the inter-regional support feasibility and the known regional distribution difficulty, the distribution tasks are graded, where this division process will consider multiple factors such as the urgency of the task and the difficulty coefficient, thereby obtaining the distribution task allocation sequence, and normalizing the distribution indicators in this allocation sequence. By establishing a unified evaluation standard, a matrix reflecting the difficulty of distribution tasks in each region is finally obtained. For example, the evaluation dimension weight of a certain region is 0.8, the support feasibility is 0.7, the difficulty adjustment coefficient is 0.3, and the distribution difficulty is 0.6. The difficulty matrix is ​​used to construct the formula: ; Among them, DTM is the delivery task difficulty matrix, To evaluate the dimension weights, SF is the support feasibility, γ is the difficulty adjustment coefficient, and CD is the regional delivery difficulty. The difficulty matrix is ​​used to construct the formula to calculate the task difficulty value: 0.8×30.5×0.7×(1+0.3×0.6)=14.8.

[0031] 104. Perform distribution time series decomposition calculation on the distribution task difficulty matrix to obtain distribution demand forecast value, and perform distribution route planning calculation on the distribution demand forecast value to obtain order delivery time interval; In this embodiment, the distribution task difficulty matrix is ​​divided into multiple time periods of order quantity to obtain a time period distribution demand table, and the time period distribution demand table and the regional distribution pressure value are weighted to obtain the regional time-varying demand; based on the historical distribution data corresponding to the area to be allocated, the regional time-varying demand is compared with the distribution trend and associated with the distribution difficulty to obtain a distribution difficulty correction value, and the distribution difficulty correction value is matched with the regional transportation capacity status to obtain the resource demand forecast; the resource demand forecast is subjected to threshold constraints and the distribution status of each distribution sub-region to obtain a prediction benchmark for each region, and the prediction benchmark for each region is checked with the regional access time to obtain the actual deliverable quantity; based on the distribution resource status data, the regional distribution demand value corresponding to the actual deliverable quantity is calculated, and the regional distribution demand value is weighted to obtain The distribution demand forecast value is obtained by combining the distribution demand forecast value with the regional access time to obtain the regional shortest travel time, and matching the regional shortest travel time with the regional distribution radius to obtain the basic distribution range; the basic distribution range is associated with the regional transportation capacity status to obtain the coverable distribution points, and the distribution distances between the coverable distribution points are calculated to obtain the distribution point time matrix; the distribution point time matrix is ​​combined by paths to obtain feasible distribution routes, and the feasible distribution routes and the regional distribution pressure value are weighted to obtain the time compensation coefficient; the time compensation coefficient is corrected with the actual distance corresponding to the feasible distribution route to obtain the distribution route duration, and the distribution route duration is matched with the order service level to obtain the time allocation sequence; the time allocation sequence is divided into intervals to obtain the order delivery time interval.

[0032] In practical applications, the delivery task difficulty matrix is ​​first divided into 30-minute basic time units. By analyzing the order distribution in each time period, a detailed time period delivery demand table is formed, and this demand table is weighted and combined with the regional delivery pressure value calculated previously. This weighted process takes into account the differences in delivery pressure in different regions. The weighted calculation formula is: ; Among them, DTR(i,t) is the time-varying demand of area i in time period t, ODM(i,t) is the original order volume of area i in time period t, RP(i) is the distribution pressure value of area i, α, β are the weight coefficients of order volume and pressure value, λ is the time attenuation coefficient, t-t0 is the difference with the reference time, so as to calculate the regional time-varying demand that more accurately reflects the dynamic changes of regional distribution demand (for example, assuming that the original order volume of a certain area from 9:00 to 9:30 is 100, the distribution pressure value is 0.8, α=0.6, β=0.4, λ=0.1, then the time-varying demand of this period is about 96.3); after obtaining the regional time-varying demand, based on the historical distribution data of the area to be allocated, the distribution trend analysis is carried out, and the difference between the historical data of the same period and the current distribution demand is compared, and the known distribution difficulty factors are combined to obtain the correction value of the distribution difficulty (wherein, this correction value not only reflects the changing trend of the distribution demand, but also takes into account the various difficulty factors that may be encountered in the actual distribution process). This correction value is then matched and analyzed with the existing capacity status of the region, and a more accurate resource demand forecast is calculated by detecting whether the corresponding data between the two are correlated and matched; the resource demand forecast is then subjected to threshold constraint processing, by setting maximum and minimum demand thresholds, and taking into account the differences in distribution status between distribution sub-regions, a balanced calculation is performed to obtain a forecast benchmark for each region, and this forecast benchmark is calibrated with the previously calculated regional access time. This calibration process verifies the feasibility of the forecast result, and ultimately obtains the actual executable distribution volume; then, based on the distribution resource status data, including information such as the number of available distribution vehicles and the working hours of the delivery personnel, the regional distribution demand value corresponding to the actual deliverable volume is calculated, and under the premise of considering the actual carrying capacity of the distribution resources and the service quality requirements, the demand values ​​of different regions are weighted to ultimately obtain an accurate distribution demand forecast value (this forecast value not only reflects the actual distribution demand of each region, but also takes into account the limitations of distribution resources and the service quality requirements).

[0033] Secondly, the distribution demand forecast value and the regional access time are combined for analysis. By considering the actual road conditions and distribution of distribution demand in the region, the shortest travel time between the distribution points in the region is calculated, and this shortest travel time is matched with the known regional distribution radius. By comprehensively considering the constraints of both time and distance, the basic distribution range that meets the distribution timeliness requirements is determined. This range is then correlated with the regional transportation capacity status for analysis. Considering factors such as the number of currently available distribution vehicles and the working hours of the delivery personnel, the actual set of distribution points that can be covered is obtained, and the actual distribution distances between these distribution points that can be covered are calculated, thereby converting these distances into distribution points. The delivery time is calculated to form a complete delivery point time matrix (where this matrix records the time cost between delivery points in detail); then based on the delivery point time matrix, multiple feasible delivery routes are generated by considering the time relationship and sequence constraints between each delivery point, and these feasible delivery routes are weighted with the previously calculated regional delivery pressure values. This process will take into account the differences in delivery pressure in different regions, so as to obtain a time compensation coefficient that reflects the actual delivery difficulty. For example, if the original delivery time is 18 minutes, the pressure coefficient is 0.2, the current pressure value is 80, and the capacity threshold is 100, the delivery routes are combined and the time compensation is calculated through the dynamic programming algorithm. The time compensation formula is: ; in, For the delivery time after compensation, is the original delivery time, γ represents the pressure coefficient, Indicates the current regional distribution pressure value. It represents the regional delivery capacity threshold, so that the compensated time can be calculated to be approximately 21.6 minutes. Then, using this time compensation coefficient, the actual distance of the feasible delivery route is corrected and calculated to obtain a more accurate delivery route duration (wherein, this duration not only takes into account the basic driving time, but also includes the additional time cost caused by delivery pressure), and the corrected delivery route duration is matched with the order service level. By considering the delivery timeliness requirements of different service levels, a detailed time allocation sequence is formed. Then, this time allocation sequence is reasonably divided into intervals. This division process will consider multiple factors such as the timeliness requirements of the order, the availability of delivery resources, and the traffic conditions of the road network, and finally a scientific and reasonable order delivery time interval is obtained.

[0034] 105. Perform delivery conflict detection on the order delivery time interval to obtain an initial delivery time window plan, and perform secondary delivery time adjustment on the initial delivery time window plan to obtain a final delivery time allocation plan.

[0035] In this embodiment, the overlapping time periods of the order delivery time intervals are counted to obtain time conflict points, and the time conflict points are sorted according to the order service level to obtain a priority processing sequence; the priority processing sequence is compared with the time period delivery capacity value to obtain the time period load value, and the time period load value is associated with the regional delivery pressure value to obtain a time adjustment range; the orders within the time adjustment range are reordered to obtain an adjusted time series, and the adjusted time series is verified with the regional transportation capacity status to obtain a regional delivery plan; the regional delivery plan is integrated according to the delivery route to obtain to the time window initial plan; obtain the real-time execution status corresponding to the time window initial plan, and compare the real-time execution status with the order delivery time interval to obtain the time deviation value; identify the orders whose delay risk values ​​exceed the threshold, obtain the order set to be adjusted, and based on the delivery resource status data, reallocate the order set to be adjusted and adjust the sub-delivery areas to obtain the regional cooperation plan; integrate the regional cooperation plan with the time window initial plan to obtain the time window adjustment plan, verify the feasibility of the time window adjustment plan and update it in real time to obtain the final delivery time allocation plan.

[0036] In practical applications, we first conduct an overlap analysis on the order delivery time intervals, that is, by scanning all orders in the same time period, we calculate the number of order overlaps at each time point, and find out the key time points where there are delivery time conflicts. We then prioritize these time conflict points according to the service levels of the orders involved, and orders with high service levels will be given priority processing rights, thus forming a priority processing sequence sorted by service importance. We then compare it in detail with the previously calculated time period deliverable capacity value, and calculate the ratio of the number of orders in each time period to the delivery capacity to obtain the time period load value reflecting the delivery pressure. We then correlate this load value with the known regional delivery pressure value, that is, by comprehensively considering the delivery pressure in the time and space dimensions, we determine the time period that each order can be delivered to. The time range of adjustment must ensure both the quality of delivery service and the rational use of delivery resources. Then, based on the determined time adjustment range, the orders within the range are reordered. This sorting process will take into account multiple factors such as the order's service level, delivery distance, and time urgency, thereby obtaining a new time series arrangement. This adjusted time series is then matched and verified with the region's current capacity status to ensure that the delivery task volume in each time period does not exceed the actual delivery capacity, thereby forming an executable regional delivery plan. Then, the regional delivery plan is integrated according to the optimized delivery route. This integration process will take into account the continuity of the route and the delivery efficiency, and ultimately form an initial time window plan that balances delivery efficiency and service quality by reasonably arranging the delivery sequence and time.

[0037] Secondly, the execution of the initial plan of the time window is monitored in real time. The actual delivery status of each order, including the order collection time, in-transit status and estimated arrival time, is obtained, and these real-time statuses are compared with the original order delivery time interval. The difference between the actual delivery time and the planned time is calculated to obtain the time deviation value reflecting the degree of delivery delay. When it is found that the time deviation value exceeds the preset threshold, the orders with delay risks are immediately identified and marked, and classified into the set of orders to be adjusted. Based on the current distribution resource status data, including the number of available distribution vehicles, the working status of the delivery personnel and the remaining working time, the set of orders to be adjusted is reallocated. The reallocation process takes into account the adjacent distribution sub-orders. Based on the resource status of the region, a new regional cooperation plan is formed by adjusting the distribution area division and reallocating distribution resources to alleviate the distribution pressure in the local area; the newly generated regional cooperation plan is then integrated with the original time window initial plan. This integration process will take into account multiple factors such as order service level, distribution route continuity and resource utilization efficiency to form a new time window adjustment plan; the feasibility of this adjustment plan is then verified by simulation calculation to verify whether it meets various distribution constraints, and it is continuously updated and optimized according to the real-time distribution status, and finally a final distribution time allocation plan that can ensure both distribution efficiency and service quality is formed. This plan can dynamically adapt to changes in the distribution environment and ensure the stable operation of the distribution system.

[0038] In the embodiment of the present invention, the application obtains standardized distribution data by multi-source collection and anomaly detection of logistics order data, distribution resource data, road network data and environmental data of the distribution area, and then extracts spatiotemporal features and calculates order priority for the acquired data to obtain the order service level; then obtains the distribution task difficulty matrix and performs time series decomposition through regional density calculation and task similarity matching; finally, path planning and conflict detection are performed based on distribution demand forecast, and the final time window allocation plan is output. Through hierarchical data processing and feature analysis, dynamic and accurate allocation of logistics distribution time windows is achieved, the dynamic characteristics and regional differences of the distribution system are fully considered, and the allocation efficiency is effectively improved; and the multi-level feature association and task evaluation strategy are adopted, which not only realizes the reasonable allocation of distribution resources, but also enhances the dynamic adjustment ability of the time window; in addition, through demand forecasting and solution optimization, the distribution efficiency and service quality are accurately balanced, thereby realizing the efficient dynamic allocation of logistics distribution time windows as a whole.

[0039] The above describes the method for dynamically allocating a time window for logistics distribution in an embodiment of the present invention. The following describes the device for dynamically allocating a time window for logistics distribution in an embodiment of the present invention. Figure 2In one embodiment of the present invention, a device for dynamically allocating a time window for logistics distribution includes: The data preprocessing module 201 is used to obtain multi-source distribution data of the area to be distributed, and perform interquartile range anomaly detection and time series normalization on the multi-source distribution data to obtain standard multi-source distribution data; The feature extraction module 202 is used to associate the multi-period order feature dimensions of the standard multi-source delivery data to obtain the order spatiotemporal distribution features, and calculate the order priority based on the order spatiotemporal distribution features to obtain the order service level; The task evaluation module 203 is used to calculate the density of the area to be assigned to the order service level to obtain the regional distribution pressure value, and to perform similarity matching of the distribution tasks on the regional distribution pressure value to obtain the distribution task difficulty matrix; The demand analysis module 204 is used to perform a delivery time series decomposition calculation on the delivery task difficulty matrix to obtain a delivery demand forecast value, and perform a delivery path planning calculation on the delivery demand forecast value to obtain an order delivery time interval; The solution generation module 205 is used to perform delivery conflict detection on the order delivery time interval to obtain an initial delivery time window solution, and perform secondary delivery time adjustment on the initial delivery time window solution to obtain a final delivery time allocation solution.

[0040] In the embodiment of the present invention, the application obtains standardized distribution data by multi-source collection and anomaly detection of logistics order data, distribution resource data, road network data and environmental data of the distribution area, and then extracts spatiotemporal features and calculates order priority for the acquired data to obtain the order service level; then obtains the distribution task difficulty matrix and performs time series decomposition through regional density calculation and task similarity matching; finally, path planning and conflict detection are performed based on distribution demand forecast, and the final time window allocation plan is output. Through hierarchical data processing and feature analysis, dynamic and accurate allocation of logistics distribution time windows is achieved, the dynamic characteristics and regional differences of the distribution system are fully considered, and the allocation efficiency is effectively improved; and the multi-level feature association and task evaluation strategy are adopted, which not only realizes the reasonable allocation of distribution resources, but also enhances the dynamic adjustment ability of the time window; in addition, through demand forecasting and solution optimization, the distribution efficiency and service quality are accurately balanced, thereby realizing the efficient dynamic allocation of logistics distribution time windows as a whole.

[0041] above Figure 2 The logistics distribution time window dynamic allocation device in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The logistics distribution time window dynamic allocation device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0042] Figure 33 is a schematic diagram of the structure of a logistics distribution time window dynamic allocation device provided by an embodiment of the present invention. The logistics distribution time window dynamic allocation device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage devices) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 can be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the logistics distribution time window dynamic allocation device 300. Furthermore, the processor 310 may be configured to communicate with the storage medium 330 to execute a series of instruction operations in the storage medium 330 on the logistics distribution time window dynamic allocation device 300.

[0043] The logistics distribution time window dynamic allocation device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that Figure 3 The structure of the logistics distribution time window dynamic allocation device shown does not constitute a limitation of the logistics distribution time window dynamic allocation device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0044] The present invention also provides a device for dynamically allocating time windows for logistics distribution. The computer device includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes each step of the method for dynamically allocating time windows for logistics distribution in the above-mentioned embodiments.

[0045] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes each step of the method for dynamic allocation of logistics distribution time windows.

[0046] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0047] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0048] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0049] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamically allocating time windows for logistics distribution, characterized in that: The method for dynamically allocating logistics distribution time windows includes: Acquire multi-source distribution data of the area to be allocated, and perform interquartile range anomaly detection and time series normalization on the multi-source distribution data to obtain standard multi-source distribution data; The standard multi-source delivery data is dimensionally associated with multi-period order features to obtain order spatiotemporal distribution features, and order priority is calculated based on the order spatiotemporal distribution features to obtain order service levels; Performing density calculation of the area to be assigned for the order service level to obtain a regional distribution pressure value, and performing similarity matching of distribution tasks for the regional distribution pressure value to obtain a distribution task difficulty matrix; Performing a delivery time series decomposition calculation on the delivery task difficulty matrix to obtain a delivery demand forecast value, and performing a delivery path planning calculation on the delivery demand forecast value to obtain an order delivery time interval; A delivery conflict detection is performed on the order delivery time interval to obtain an initial delivery time window plan, and a secondary delivery time adjustment is performed on the initial delivery time window plan to obtain a final delivery time allocation plan.

2. The method for dynamically allocating logistics distribution time windows according to claim 1 is characterized in that: The performing of interquartile range anomaly detection and time series normalization on the multi-source distribution data to obtain standard multi-source distribution data includes: Calculate the interquartile range of order delivery time for the multi-source delivery data to obtain a normal order time interval, and determine a set of abnormal orders corresponding to the time period deviation in the normal order time interval; Based on the historical order time of the area to be assigned, the delivery time of the abnormal order set is compared to obtain a high-frequency delivery time period, and the time deviation between the high-frequency delivery time period and each abnormal order in the abnormal order set is calculated to obtain a time adjustment interval; The delivery time period capacity is allocated to the delivery orders corresponding to the time adjustment interval to obtain multi-source delivery data after the delivery order time is adjusted, and based on a preset maximum delivery threshold per unit time, the delivery time demand is divided into multiple time periods and the capacity constraint is calculated for the multi-source delivery data after the delivery order time is adjusted to obtain standard multi-source delivery data.

3. The method for dynamically allocating logistics distribution time windows according to claim 1 is characterized in that: The standard multi-source distribution data includes logistics order data, distribution resource status data, distribution road network data and distribution environment data. The multi-period order feature dimension association is performed on the standard multi-source distribution data to obtain the order spatiotemporal distribution features, including: Based on the area to be allocated, the logistics order data is grouped into delivery areas according to the order quantity of the time period to obtain the regional order distribution, and based on the distribution road network data, the route distance of the regional order distribution is calculated to obtain the regional distribution radius; Based on the distribution resource status data, determine the number of delivery personnel corresponding to the regional distribution radius, obtain the regional transportation capacity distribution status, and divide the regional transportation capacity distribution status into time periods to obtain the delivery capacity value of the time period; Based on the delivery environment data, the delivery capacity value of the time period is matched with the delivery road network to obtain the regional access time, and the regional access time is aggregated for time periods to obtain the spatiotemporal distribution characteristics of the order.

4. The method for dynamically allocating logistics distribution time windows according to claim 3 is characterized in that: The step of calculating the order priority based on the temporal and spatial distribution characteristics of the order to obtain the order service level includes: Determine the time period order density corresponding to the unit time order quantity in the order spatiotemporal distribution characteristics, and perform ratio calculation on the time period order density based on the maximum delivery quantity of the delivery vehicle in the delivery resource status data to obtain the time period load rate; Based on the delivery capacity value of the time period, weighted calculation is performed on the load rate of the time period to obtain a delivery load index, and the delivery load index is accumulated in the delivery target area to obtain the regional time period pressure; Based on the preset route distance coefficient and the delivery vehicle data in the delivery resource status data, the delivery guarantee rate corresponding to the regional time period pressure is calculated, and based on the delivery environment data, the delivery guarantee rate is subjected to secondary delivery calibration to obtain the regional service capacity; Based on a preset regional saturation index, the regional service capacity is divided into multiple delivery levels and weighted averaged to obtain an initial order level, and based on the maximum delivery capacity corresponding to the area to be allocated, the initial order level is calibrated to obtain an order service level.

5. The method for dynamically allocating logistics distribution time windows according to claim 3 is characterized in that: The density calculation of the area to be allocated for the order service level to obtain the regional distribution pressure value includes: Combining the order service level and the order spatiotemporal distribution characteristics with the delivery area characteristics to obtain a regional priority delivery sequence, and matching the regional priority delivery sequence with the regional delivery radius to obtain a delivery coverage range; Calculate the ratio of the number of delivery orders within the delivery coverage area to the regional transportation capacity distribution status to obtain the regional task density, and combine the regional task density with the regional access time to obtain the regional delivery difficulty; The regional distribution difficulty is associated with adjacent distribution sub-regions to obtain an inter-region coordination coefficient, and the distribution resources are summarized and calculated for the inter-region coordination coefficient to obtain a regional distribution pressure value.

6. The method for dynamically allocating logistics distribution time windows according to claim 5 is characterized in that: The similarity matching of the delivery tasks for the regional delivery pressure values ​​to obtain the delivery task difficulty matrix includes: Accumulating the regional distribution pressure value and the order volume of adjacent regions in the regional order distribution to obtain the regional cluster order volume, and performing regional distribution resource comparison on the regional cluster order volume based on the distribution resource status data to obtain the cluster distribution capacity difference; The cluster distribution capacity differences are sorted by regional access time to obtain regional distribution urgency, and the regional distribution urgency is combined with the regional access time to obtain inter-regional support feasibility; The inter-regional support feasibility and regional distribution difficulty are graded to obtain a distribution task allocation sequence, and the distribution index of the distribution task allocation sequence is normalized to obtain a distribution task difficulty matrix.

7. The method for dynamically allocating logistics distribution time windows according to claim 3 is characterized in that: The step of performing distribution time series decomposition calculation on the distribution task difficulty matrix to obtain a distribution demand forecast value includes: The distribution task difficulty matrix is ​​divided into multiple time period order quantities to obtain a time period distribution demand table, and the time period distribution demand table and the regional distribution pressure value are weighted to obtain the regional time-varying demand quantity; Based on the historical distribution data corresponding to the area to be allocated, the distribution trend comparison and distribution difficulty correlation of the time-varying demand of the area are performed to obtain a distribution difficulty correction value, and the distribution difficulty correction value is matched with the regional transportation capacity status to obtain a resource demand forecast; The resource demand forecast is subjected to threshold constraints and the distribution status of each distribution sub-region is balanced to obtain the forecast benchmark of each region, and the forecast benchmark of each region is verified with the regional access time to obtain the actual deliverable quantity; Based on the distribution resource status data, the regional distribution demand value corresponding to the actual deliverable quantity is calculated, and the regional distribution demand value is weighted to obtain a distribution demand forecast value.

8. A device for dynamically allocating time windows for logistics distribution, characterized in that: The logistics distribution time window dynamic allocation device comprises: A data preprocessing module is used to obtain multi-source distribution data of the area to be distributed, and perform interquartile range anomaly detection and time series normalization on the multi-source distribution data to obtain standard multi-source distribution data; A feature extraction module is used to associate the multi-period order feature dimensions of the standard multi-source delivery data to obtain the order spatiotemporal distribution features, and to calculate the order priority based on the order spatiotemporal distribution features to obtain the order service level; A task evaluation module is used to calculate the density of the area to be assigned to the order service level to obtain the regional distribution pressure value, and to perform similarity matching of the distribution tasks on the regional distribution pressure value to obtain the distribution task difficulty matrix; A demand analysis module is used to perform a delivery time series decomposition calculation on the delivery task difficulty matrix to obtain a delivery demand forecast value, and perform a delivery route planning calculation on the delivery demand forecast value to obtain an order delivery time interval; The solution generation module is used to perform delivery conflict detection on the order delivery time interval to obtain an initial delivery time window solution, and perform secondary delivery time adjustment on the initial delivery time window solution to obtain a final delivery time allocation solution.

9. A logistics distribution time window dynamic allocation device, characterized in that: The logistics distribution time window dynamic allocation device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory so that the logistics distribution time window dynamic allocation device executes each step of the logistics distribution time window dynamic allocation method as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the various steps of the method for dynamic allocation of logistics distribution time windows as described in any one of claims 1-7 are implemented.

Images