A method for intelligent order scheduling

Abstract

This invention discloses an intelligent order scheduling method, belonging to the field of order scheduling management technology. The method includes acquiring and analyzing order information to be scheduled, obtaining the demand quantity, time gradient data, and gradient purchase quantity of the goods to be purchased; acquiring the production efficiency, current inventory, and delivery time of each supplier, and determining the satisfaction level of each supplier with the goods to be purchased based on the time gradient data, thus obtaining a first supplier to be allocated and a second supplier to be allocated; for the first supplier to be allocated, matching and calculating the demand quantity of the goods to be purchased with a supplier quantity-price comparison table, and selecting the first procurement plan with the lowest purchase price; for the second supplier to be allocated, performing combination judgment and price analysis on the suppliers to obtain the second procurement plan with the lowest purchase price; and comparing the first and second procurement plans to obtain the optimal allocation plan. This invention has the effect of reducing procurement costs.

Description

Technical Field

[0001] This application relates to the field of order scheduling management technology, and in particular to an intelligent order scheduling method. Background Technology

[0002] In the current retail and supply chain management field, the efficiency of order processing and delivery between stores and suppliers directly affects the operational efficiency of the entire supply chain and customer satisfaction. With increasingly fierce market competition and diversified consumer demands, traditional supply chain order scheduling methods are no longer sufficient to meet the high-efficiency and precise needs of modern commerce.

[0003] In related technologies, under the traditional model, the allocation and scheduling of store orders mainly rely on human experience and judgment. This process is not only time-consuming and labor-intensive, but also prone to errors. Especially when faced with large-scale order volumes and complex delivery requirements, the limitations of manual scheduling are particularly evident, making it difficult to achieve optimal resource allocation.

[0004] Existing scheduling methods often only consider the supply factors of a single supplier (such as cost, supply inventory, etc.) when making decisions, while ignoring the possibility of coordinated scheduling of multiple suppliers. This may result in suboptimal scheduling solutions that fail to achieve a balance and optimization of multiple objectives such as delivery costs, order fulfillment rates, and supplier workload, and therefore requires improvement. Summary of the Invention

[0005] To achieve accurate matching and intelligent scheduling of supplier resources, this application provides an intelligent order scheduling method.

[0006] This application provides an intelligent order scheduling method, which adopts the following technical solution: An intelligent order scheduling method includes: Obtain the order information to be scheduled, and perform statistical analysis on the product information of the goods to be purchased in the order information to obtain the goods to be purchased and their demand. Based on the order information, determine the reserved time slots for each order to purchase goods. Based on the category of goods to be purchased, statistically analyze the reserved time slots for the same category of goods to be purchased to obtain time gradient data, and obtain the gradient purchase quantity of each goods to be purchased under the corresponding time gradient. Obtain the production efficiency, current inventory, and delivery time of each supplier, and combine this with the time gradient data of each product to be purchased to determine the satisfaction level of each supplier for each product to be purchased, thereby obtaining the first supplier to be allocated that can fully satisfy the needs and the second supplier to be allocated that cannot fully satisfy the needs. For the first supplier to be allocated, the demand for the goods to be purchased is matched and calculated with the quantity and price comparison table of each supplier in the first supplier to be allocated, so as to obtain the purchase price of each supplier, and the purchase price is compared to select the first purchase plan with the lowest purchase price. For the second supplier to be allocated, the supplier is combined and price is analyzed based on the corresponding supplier’s production efficiency, current inventory, delivery time, time gradient data of the goods to be purchased and the gradient purchase quantity under the corresponding time gradient to obtain the second purchase plan with the minimum purchase price. The prices of the first and second procurement plans are compared, and the procurement plan with the lowest procurement price is taken as the optimal allocation plan. The corresponding purchase order is then generated and sent to the corresponding supplier.

[0007] Preferably, the order information to be scheduled is obtained, and the order information submitted by the same store is classified and the purchase quantity is counted based on the store information and built-in product categories to obtain the category purchase quantity of the same product category. Based on the store information, obtain the historical category purchase volume and corresponding inventory information of the corresponding store, judge the historical category purchase volume and corresponding inventory information, determine the dynamic balance relationship between the historical category purchase volume and the corresponding inventory information, and record it as the first balance relationship. Obtain the current store's inventory information, and compare the current store's inventory information with the category purchase quantity in the submitted order information to determine the dynamic balance relationship between the current category purchase quantity and the corresponding inventory information, and record it as the second balance relationship; The first balance relationship is compared with the second balance relationship. If the second balance relationship is greater than the first balance relationship, the order information submitted by the store is determined to be abnormal.

[0008] Preferably, when it is determined that there is an anomaly in the order information submitted by the store in the order information to be dispatched, the daily inventory log of each product to be purchased in the order information under the corresponding product category is obtained, and the consumption status of each product to be purchased is determined based on the daily inventory log. Based on consumption data, forecast the consumption of goods for the next procurement cycle to obtain the expected consumption data. Based on the expected consumption data and the current inventory level, determine the expected procurement quantity. Based on the product information of the goods to be purchased, the expected purchase quantity is compared with the purchase demand quantity in the order information. If the expected purchase quantity is greater than the purchase demand quantity, it is determined that the purchase demand quantity of the goods is normal. If the expected purchase quantity is less than the purchase demand quantity, the purchase demand quantity of the product is determined to be abnormal, and an alarm signal for abnormal purchase demand of the product is output.

[0009] Preferably, the production time is obtained by calculating the difference between the reserved gap time in each time gradient stage and the delivery time of each supplier. Based on production duration and production efficiency, determine the quantity of goods that the supplier can produce under the corresponding time gradient, and obtain the production quantity. Compare the sum of production volume and the corresponding supplier's current inventory with the gradient purchase volume of the corresponding time gradient data; If the sum of the production volume and the current inventory of the corresponding supplier is greater than the gradient purchase volume of the corresponding time gradient data, then it is determined that the supplier can meet the corresponding purchase volume. Based on the suppliers' ability to meet the requirements of the purchased goods, suppliers who can meet the corresponding purchase volume of the purchased goods in each time gradient are designated as the first suppliers to be allocated, and suppliers who can only meet the purchase volume in some time gradients are designated as the second suppliers to be allocated.

[0010] Preferably, when it is determined that some suppliers cannot fully meet the tiered procurement volume under each time gradient, the tiered procurement volume of the unmet time gradient is obtained and recorded as the unmet tiered procurement volume. Based on time gradient data and unmet time gradients, obtain the gradient purchase quantity of the time gradient before the unmet time gradient time point, and record it as the met gradient purchase quantity. The difference between the tiered procurement quantity to be satisfied and the tiered procurement quantity already satisfied is calculated to obtain the compensation difference. Based on the compensation difference, the difference in the gradient procurement quantity of the time gradient located after the time gradient point that cannot be met is calculated to obtain the corrected comparison value. The corrected comparison value is compared with the sum of the supplier's production and current inventory under the corresponding time gradient until all time gradients have been compared. Suppliers are marked based on the time gradients that cannot be met, and a second supplier with a time gradient gap is obtained.

[0011] Preferably, based on the supplier's production efficiency, current inventory, delivery time, and time gradient data of the goods to be purchased, the maximum gradient purchase quantity that each supplier in the second list of suppliers to be allocated can meet is determined and recorded as the maximum supply quantity of the corresponding supplier. The difference between the maximum supply and the demand for the goods to be purchased is calculated to obtain the procurement difference data. Based on the procurement difference data and the maximum supply of each supplier, other suppliers other than the supplier are matched until the sum of the maximum supply of the combined suppliers is greater than the demand for the goods, thus obtaining the combined data. Select a set of combinations from the combined data, construct a cost function with conditional constraints, transform the cost function using a mixed integer nonlinear programming algorithm, and solve the transformed cost function using a business solver to obtain the optimal solution for the corresponding combination. The combination allocation result under the corresponding optimal solution is recorded as the combination allocation scheme, and the purchase price of the corresponding combination allocation result is recorded as the combination allocation price. Based on the combined data, the combined allocation prices under each combined case are compared with each other, and the combined allocation scheme and combined allocation price with the smallest purchase price in the combined data are selected. The result of this selection is recorded as the second purchase scheme.

[0012] Preferably, the daily inventory logs of each product in each store are obtained, and the sales volume of each product is judged based on the daily inventory logs to determine the sales situation of each product in each store. Based on the built-in product categories, the sales performance of the same type of product is analyzed discretely to determine whether some products are selling worse than others. If it is determined that the sales performance of some products is worse than that of other products, then these products are recorded as products to be evaluated. The sales performance of the corresponding products to be evaluated in each store is obtained, and the sales performance of the products to be evaluated is discretely judged to determine whether there are some stores whose sales performance is better than that of other stores. If it is determined that some stores are performing better than others, then the inventory of the products to be evaluated in the underperforming stores will be adjusted based on the sales performance of the top-performing stores. If it is determined that there are no stores where sales are better than other stores, then obtain the supplier information of the product to be evaluated, and read the corresponding supplier's purchase order based on the supplier information of the product to be evaluated. The corresponding supplier's purchase order is matched with the product to be evaluated. If the proportion of the product to be evaluated in the corresponding purchase order is greater than the built-in evaluation threshold, the supplier is deemed unqualified. If the proportion of the product to be evaluated in the corresponding purchase order is less than the built-in evaluation threshold, the supplier will be marked with a product based on the product to be evaluated, and information on suppliers with poor product markings will be obtained.

[0013] In summary, this application includes at least one of the following beneficial technical effects: 1. By acquiring and analyzing order information, we can quickly identify the goods to be purchased and their demand, reducing the time and error rate of manual statistics and improving the efficiency of procurement preparation. By constructing time gradient data based on the urgency of each product's demand, and arranging procurement plans according to this data, we ensure the rationality and timeliness of the procurement plans. By comprehensively evaluating suppliers' production efficiency, inventory levels, and delivery times, and combining this with time gradient data, we can accurately distinguish between suppliers that can fully meet the demand and those that cannot. This allows us to make single supplier judgments for suppliers that can fully meet the demand and multiple supplier combination judgments for those that cannot, thereby determining the procurement cost under the corresponding circumstances. By selecting the procurement plan with the lowest cost, we can effectively reduce procurement costs, while the backup of multiple plans ensures the stability and reliability of the supply chain. 2. By classifying and statistically analyzing the order information submitted by the same store, refined management of order data is achieved, providing a data foundation for subsequent steps. A first balance relationship is constructed by obtaining historical purchase volumes and corresponding inventory information for each product category. Simultaneously, a second balance relationship is constructed based on the current category purchase volume and inventory information, providing a scientific basis for identifying order anomalies. By comparing the first and second balance relationships, abnormal order information can be quickly identified, helping companies take timely measures to avoid inventory backlog. Furthermore, once anomalies are identified, daily inventory logs are analyzed to predict product consumption for the next procurement cycle. By comparing the projected purchase volume with the actual purchase demand in the order information, the rationality of the purchase demand is further verified, and timely anomaly alarm signals are output to help companies adjust their procurement strategies and avoid inventory backlog or shortages. 3. By comprehensively utilizing supplier production efficiency, current inventory levels, delivery times, and time-series data of goods to be procured, the maximum supply of each supplier in the second batch of suppliers to be allocated is determined. Then, by calculating the difference between the maximum supply and the demand for the goods to be procured, the quantity of goods that the supplier itself cannot meet is determined (procurement difference data). Then, by matching other suppliers with this unmet quantity of goods, suppliers that can meet the procurement difference data are combined with the supplier to obtain the corresponding combination data. By allocating the procurement quantity for a single combination in the combination data, the procurement plan with the lowest procurement price is obtained. Finally, the procurement prices of all the procurement plans under all the combination cases in the combination data are compared to further confirm the procurement plan with the optimal procurement price in the combination data, ensuring the accuracy of the optimal price determination. Attached Figure Description

[0014] Figure 1 This is a flowchart of the steps of the intelligent order scheduling method in this embodiment. Detailed Implementation

[0015] The following is in conjunction with the appendix Figure 1 This application will be described in further detail.

[0016] This application discloses an intelligent order scheduling method.

[0017] Example: Figure 1 As shown, the present invention provides an intelligent order scheduling method, comprising: S1, obtain the order information to be scheduled, and perform statistical judgment on the product information of the goods to be purchased in the order information to obtain the goods to be purchased and their demand quantity; S2, determine the reserved time slot for each order to purchase goods based on the order information, and statistically analyze the reserved time slot for the same type of goods according to the category of goods to be purchased to obtain time gradient data, and obtain the gradient purchase quantity of each goods to be purchased under the corresponding time gradient; where reserved time slot refers to the latest delivery time of the goods, for example, goods A need to be delivered within 7 days.

[0018] S3: Obtain the production efficiency, current inventory, and delivery time of each supplier, and combine them with the time gradient data of each product to be purchased to determine the satisfaction of each supplier with each product to be purchased, and obtain the first supplier to be allocated that can be fully satisfied and the second supplier to be allocated that cannot be fully satisfied. S4. For the first supplier to be allocated, match and calculate the demand for the goods to be purchased with the quantity and price comparison table of each supplier in the first supplier to be allocated, obtain the purchase price of each supplier, compare the purchase prices, and select the first purchase plan with the smallest purchase price. S5. For the second supplier to be allocated, based on the corresponding supplier's production efficiency, current inventory, delivery time, time gradient data of the goods to be purchased, and the gradient purchase quantity under the corresponding time gradient, perform combination judgment and price analysis on the supplier to obtain the second purchase plan with the minimum purchase price. S6. Compare the prices of the first procurement plan and the second procurement plan, select the procurement plan with the lowest procurement price as the optimal allocation plan, generate the corresponding purchase order, and send the purchase order to the corresponding supplier.

[0019] In this embodiment, by acquiring and analyzing order information, the goods to be purchased and their demand quantities can be quickly calculated, reducing the time and error rate of manual statistics and improving the efficiency of procurement preparation. By constructing time gradient data using the urgency of each product's demand, and arranging the procurement plan based on the time gradient data, the rationality and timeliness of the procurement plan are ensured. By comprehensively evaluating the supplier's production efficiency, inventory, and delivery time, and combining the time gradient data, suppliers that can fully meet the demand and those that cannot are accurately distinguished. Thus, a single supplier judgment is made for suppliers that can fully meet the demand, and a multi-supplier combination judgment is made for suppliers that cannot fully meet the demand, thereby determining the procurement cost under the corresponding circumstances. By selecting the procurement plan with the lowest cost, the procurement cost is effectively reduced, while the stability and reliability of the supply chain are ensured by using multiple alternative plans.

[0020] For example, suppose we need to purchase three kinds of goods, A, B and C. Store 1 applies for a purchase quantity of 100 of A, Store 2 applies for a purchase quantity of 100 of A, and Store 3 applies for a purchase quantity of 100 of A. By analyzing the purchase volumes of the three stores, the total purchase volume of product A was determined to be 300. However, due to varying urgency of demand for the same product among different stores, Store 1 requires product A to arrive within 7 days, Store 2 within 15 days, and Store 3 within 24 days. Therefore, there are three time tiers for purchasing product A: 7 days, 15 days, and 24 days. These can be denoted as the first, second, and third time tiers, respectively. These three time tiers are collectively referred to as the time tier data, with the first time tier corresponding to a tier purchase volume of 100, the second time tier to 200, and the third time tier to 300.

[0021] The time gradient data indicates the latest delivery time for the corresponding purchase quantity by the supplier. That is, the supplier needs to deliver at least 100 units of product A within 7 days, at least 200 units of product A within 15 days, and at least 300 units of product A within 24 days.

[0022] After clarifying the conditions that suppliers need to meet, suppliers are matched according to these conditions. Suppliers that can meet the above conditions with a single supplier are designated as the first supplier to be assigned, and suppliers that require multiple suppliers to meet the above conditions are designated as the second supplier to be assigned.

[0023] Since different purchase quantities correspond to different unit prices, it is necessary to match the quantity-price comparison tables of various suppliers based on the purchase quantity to determine the unit price for the corresponding purchase quantity. Generally, the larger the purchase quantity, the lower the corresponding unit price, and the price follows a piecewise function distribution. For example, a purchase quantity of 0-50 corresponds to a unit price of 100; a purchase quantity of 50-80 corresponds to a unit price of 80; a purchase quantity of 80-100 corresponds to a unit price of 70, and so on.

[0024] For the first supplier to be allocated, since this type of supplier can complete this order on its own, the total demand of this order is taken as the purchase quantity of this type of supplier. The purchase price of the corresponding supplier is then calculated. By comparing prices, the supplier with the lowest cost price and the corresponding purchase quantity are taken as the best purchase plan for the first supplier to be allocated.

[0025] Since the best procurement option selected at this point may not be the lowest price, it is necessary to evaluate the second supplier to be allocated.

[0026] For the second supplier to be assigned, since this type of supplier cannot complete the order on its own, it is necessary to combine this type of supplier with other suppliers (including the first and second suppliers to be assigned) to ensure that the combined supplier can meet the order requirements.

[0027] For example, among the second suppliers to be allocated, there is a supplier whose unit price is 60 when the purchase quantity is 80-100, while the unit price of the best purchase plan of the first supplier to be allocated is 70. Therefore, it is necessary to judge this situation and determine the combined price.

[0028] By calculating and comparing the prices of the combined options, the procurement plan with the lowest procurement cost among the second batch of suppliers to be allocated is selected. For example, the procurement plan for the second batch of suppliers to be allocated could be supplier a purchasing 80 units and supplier b purchasing 20 units.

[0029] After determining the overall procurement cost and the combined procurement cost, the two are compared to select the procurement plan with the lower procurement cost as the final procurement plan. Based on the procurement plan, a corresponding purchase order is generated for the corresponding supplier to carry out the procurement.

[0030] In step S1, the order information to be scheduled is obtained, and the product information of the goods to be purchased in the order information is statistically analyzed to obtain the goods to be purchased and their demand quantities, including the following steps: S11: Obtain the order information to be scheduled, and classify and calculate the purchase volume of the order information submitted by the same store based on the store information and the built-in product categories, so as to obtain the category purchase volume of the same product category. S12, Based on the store information, obtain the historical category purchase volume and the corresponding historical category purchase volume inventory information of the corresponding store, judge the historical category purchase volume and the corresponding inventory information, determine the dynamic balance relationship between the historical category purchase volume and the corresponding inventory information, and record it as the first balance relationship; S13, obtain the current store's inventory information, and compare the current store's inventory information with the category purchase quantity in the submitted order information to determine the dynamic balance relationship between the current category purchase quantity and the corresponding inventory information, and record it as the second balance relationship; S14. Compare the first balance relationship with the second balance relationship. If the second balance relationship is greater than the first balance relationship, it is determined that the order information submitted by the store is abnormal.

[0031] S15, when it is determined that there is an anomaly in the order information submitted by the store in the order information to be dispatched, obtain the daily inventory log of each product to be purchased in the order information of the corresponding product category, and determine the consumption status of each product to be purchased based on the daily inventory log. S16, Based on the consumption situation, predict the consumption of goods for the next procurement cycle to obtain the expected consumption data, and determine the expected procurement quantity based on the expected consumption data and the current inventory level; S17. Based on the product information of the goods to be purchased, compare the expected purchase quantity with the purchase demand quantity in the order information. If the expected purchase quantity is greater than the purchase demand quantity, it is determined that the purchase demand quantity of the goods is normal. S18. If the expected purchase quantity is less than the purchase demand quantity, the purchase demand quantity of the product is determined to be abnormal, and an alarm signal for abnormal purchase demand of the product is output.

[0032] In this embodiment, by classifying and statistically analyzing the order information submitted by the same store, refined management of order data is achieved, providing a data foundation for subsequent steps. By acquiring the historical purchase volume and corresponding inventory information for the corresponding product category, a first balance relationship is constructed. Simultaneously, based on the current category purchase volume and inventory information, a second balance relationship is constructed, thus providing a scientific basis for order anomaly identification. By comparing the first and second balance relationships, abnormal order information can be quickly identified, helping enterprises to take timely measures to avoid inventory backlog. Furthermore, after anomalies are identified, the daily inventory log is analyzed to predict the consumption of goods in the next procurement cycle. By comparing the expected purchase volume with the actual purchase demand in the order information, the rationality of the purchase demand is further verified, and anomaly alarm signals are output in a timely manner to help enterprises adjust their procurement strategies and avoid inventory backlog or shortages.

[0033] For example, suppose store 1 submits an order for 100 units of product A. By obtaining the store's historical procurement information, we can determine the procurement quantity of product A and the corresponding inventory quantity during the historical procurement process. For instance, in the historical procurement process, when the procurement quantity of product A was 100, the corresponding inventory was 50; when the procurement quantity was 50, the corresponding inventory was 100; and when the procurement quantity was 80, the corresponding inventory was 70. Therefore, we can determine that the total quantity of product A remained constant (150 units) during the historical procurement process, thus establishing the first equilibrium relationship in the historical procurement process. However, the current actual procurement quantity is 100, but the inventory is 100, so the total quantity of product A is now 200, which is much larger than the 150 units in the historical procurement. Therefore, we can determine that there is a procurement anomaly for this type of product.

[0034] When a procurement anomaly is identified, it is necessary to further determine the anomaly items and, at the same time, determine the reasonableness of the procurement anomaly by identifying the anomaly items.

[0035] By obtaining the daily inventory logs for each specific item under Category A, the sales situation of each item can be determined based on the daily inventory logs. For example, if the original remaining inventory of 50 units could sell for 5 days, but when sales improve, the remaining inventory of 50 units can only sell for 2.5 days, then to ensure no stockouts occur, purchases need to be made when the remaining inventory reaches 100 units to fill the 5-day gap. Therefore, this abnormal purchasing situation is reasonable, while the opposite is not, and an alert should be triggered to facilitate decision-making.

[0036] In step S3, the production efficiency, current inventory, and delivery time of each supplier are obtained, and the satisfaction level of each supplier with each item to be purchased is determined by combining the time gradient data of each item to be purchased. This results in the first supplier that can be fully satisfied and the second supplier that cannot be fully satisfied. The steps include: S31, calculate the difference between the reserved gap time in each time gradient stage and the delivery time of each supplier to obtain the production time. S32, based on production time and production efficiency, determine the quantity of goods that the supplier can produce under the corresponding time gradient, and obtain the production quantity; S33 compares the sum of production volume and the current inventory of the corresponding supplier with the gradient purchase volume of the corresponding time gradient data; S34. If the sum of the production quantity and the current inventory quantity of the corresponding supplier is greater than the gradient purchase quantity of the corresponding time gradient data, then it is determined that the supplier can meet the corresponding purchase quantity. S35, based on the suppliers' ability to meet the requirements of the purchased goods, suppliers who can meet the corresponding purchase volume of the purchased goods in each time gradient are designated as the first suppliers to be allocated, and suppliers who can only meet the purchase volume in some time gradients are designated as the second suppliers to be allocated.

[0037] S351, when it is determined that some suppliers cannot fully meet the tiered purchase quantities under each time gradient, obtain the tiered purchase quantities of the time gradients that cannot be met, and record them as the tiered purchase quantities to be met. S352, based on time gradient data and unmet time gradients, obtain the gradient purchase quantity of the time gradient before the unmet time gradient time point, and record it as the met gradient purchase quantity. S353, calculate the difference between the tiered procurement quantity to be satisfied and the tiered procurement quantity already satisfied, and obtain the compensation difference; S354, based on the compensation difference, calculate the difference of the gradient procurement quantity for the time gradient after the time point of the unmet time gradient, and obtain the corrected comparison value. S355, compare the corrected comparison value with the sum of the supplier's production quantity and current inventory quantity under the corresponding time gradient, until all time gradients have been compared, and mark the suppliers based on the unmet time gradients to obtain the second supplier to be allocated with time gradient vacancies.

[0038] In this embodiment, the production duration of each time gradient is calculated to determine the supplier's production volume. Combined with the supplier's current inventory, the supplier's ability to meet the demand for goods in each time gradient is accurately assessed. When it is determined that a supplier's supply capacity cannot meet the gradient purchase volume of the corresponding time gradient, the gradient purchase volume of the unmet time gradient of the supplier is determined, thereby determining the quantity of goods that need to be compensated in the corresponding time gradient. The gradient purchase volume in subsequent time gradients is corrected by the compensation difference. The corrected gradient purchase volume is then used to re-evaluate the supplier, thereby determining the time gradient that the supplier can meet in the entire time gradient data. By marking suppliers with time gradient gaps, the transparency of the supply chain is enhanced. At the same time, by subdividing suppliers into first-to-be-assigned suppliers and second-to-be-assigned suppliers according to their satisfaction with each time gradient, it is helpful to adopt differentiated strategies for different categories of suppliers during the procurement process, thereby improving resource allocation efficiency and procurement flexibility.

[0039] For example, suppose we need to purchase product A. We need to purchase 100 units in the first time gradient, 200 units in the second time gradient, and 300 units in the third time gradient. The first time gradient is 7 days, the second time gradient is 15 days, and the third time gradient is 24 days.

[0040] Assuming supplier A has an existing inventory of 50 units, a production efficiency of 15 units per day, and a delivery time of 3 days, then the effective production time in the first time gradient is 4 days, resulting in a production output of 4 × 15 + 50 = 110 > 100, satisfying the first time gradient demand. In the second time gradient, the effective production time is 12 days, resulting in a production output of 12 × 15 + 50 = 230 > 200, satisfying the second time gradient demand. In the third time gradient, the effective production time is 21 days, resulting in a production output of 21 × 15 + 50 = 365 > 100, satisfying the third time gradient demand. Therefore, it can be determined that supplier A fully meets the procurement requirements for product A.

[0041] If supplier b has an existing inventory of 40 units, a production efficiency of 13 units per day, and a delivery time of 3 days, then the effective production time in the first time gradient is 4 days, and the production result is 4×13+40=92<100, which meets the demand of the first time gradient. Since the first time gradient cannot meet the demand, the compensation difference is 100. Therefore, the number of goods required to meet the demand in the second time gradient is 100, and the number of goods required to meet the demand in the third time gradient is 200.

[0042] Within the second time gradient, the effective production time is 12 days, and the production result is 12 × 13 + 40 = 196 > 100, satisfying the demand for the second time gradient. Within the third time gradient, the effective production time is 21 days, and the production result is 21 × 13 + 40 = 313 > 200, satisfying the demand for the third time gradient. Therefore, it can be determined that supplier b only satisfies the quantity of goods required for product A in both the second and third time gradients. The first time gradient of supplier b is then marked.

[0043] In step S5, for the second supplier to be allocated, based on the corresponding supplier's production efficiency, current inventory, delivery time, time gradient data of the goods to be purchased, and the gradient purchase quantity under the corresponding time gradient, the supplier is combined and price is analyzed to obtain the second procurement plan with the minimum purchase price, including the following steps: S51, based on the supplier's production efficiency, current inventory, delivery time and time gradient data of the goods to be purchased, determine the maximum gradient purchase quantity that each supplier in the second supplier to be allocated can meet, and record it as the maximum supply quantity of the corresponding supplier. S52, calculate the difference between the maximum supply and the demand for the goods to be purchased to obtain the procurement difference data, and match other suppliers that are not itself with the procurement difference data and the maximum supply of each supplier until the sum of the maximum supply of the combined suppliers is greater than the demand for the goods to be purchased, and obtain the combined data. S53. Select a set of combinations from the combined data, construct a cost function with conditional constraints, and transform the cost function using a mixed-integer nonlinear programming algorithm. Then, use a commercial solver to solve the transformed cost function to obtain the optimal solution for the corresponding combination. The combination allocation result under the corresponding optimal solution is recorded as the combination allocation scheme, and the purchase price of the corresponding combination allocation result is recorded as the combination allocation price. Among them, the mixed-integer nonlinear programming (MINLP) linearizes the piecewise function by introducing binary variables, transforming it into mixed-integer linear programming (MILP), and is solved using a commercial solver (such as Gurobi, CPLEX).

[0044] S54. Based on the combined data, the combined allocation prices under each combined case are compared with each other, and the combined allocation scheme and combined allocation price with the smallest purchase price in the combined data are selected. The result of the selection is recorded as the second purchase scheme.

[0045] In this embodiment, by utilizing the supplier's production efficiency, current inventory, delivery time, and time gradient data of the goods to be purchased, the maximum supply of each supplier in the second list of suppliers to be allocated is determined. Then, by calculating the difference between the maximum supply and the demand for the goods to be purchased, the quantity of goods that the supplier itself cannot meet (purchase difference data) is determined. Then, by matching other suppliers with the quantity of goods that cannot be met, suppliers that can meet the purchase difference data are combined with the supplier to obtain the corresponding combination data. By allocating the purchase quantity for a single combination in the combination data, the purchase plan with the lowest purchase price is obtained. Finally, the purchase prices of all the purchase plans under all the combination cases in the combination data are compared to further confirm the purchase plan with the optimal purchase price in the combination data, ensuring the accuracy of the optimal price judgment.

[0046] For example, suppose there are two time gradients, denoted as T1 and T2, where T1 is 7 days and T2 is 15 days. The procurement demand is 100 units in phase T1 and 200 units in phase T2.

[0047] Supplier A has an existing inventory of 40 units, a production efficiency of 13 units per day, and a delivery time of 3 days. Therefore, the effective production time in stage T1 is 4 days, resulting in a production quantity of 4 × 13 + 40 = 92 < 100. To meet the demand in stage T1, it is necessary to combine with other suppliers (such as supplier B) to fulfill the demand in stage T1. Since supplier A does not meet the demand in stage T1, the demand in stage T2 is 200 - 100 = 100, with an effective production time of 15 - 7 = 8 days. The production quantity in stage T2 is 8 × 13 + 40 = 144 > 100.

[0048] At this point, the maximum supply that the supplier can meet within the T1+T2 time period is 92+100=192, which is 8 units different from the procurement demand of 200. Therefore, these 8 units are matched with other suppliers.

[0049] Assuming a successful match with supplier b, the supply demand at this point is 200 units, allocated to suppliers a and b. By using the constraint that supplier a does not meet the requirements in stage T1 as a constraint, a cost function is constructed for the procurement costs of suppliers a and b. Then, a mixed-integer nonlinear partitioning algorithm is used to transform the cost function, and a business solver is used to solve the transformed cost function, thus obtaining the optimal allocation scheme for this combination. The optimal allocation schemes under each combination are then compared to further determine the final procurement scheme under the combined conditions, making the judgment more accurate.

[0050] An intelligent order scheduling method further includes the following steps: Obtain daily inventory logs for each product in each store, and determine the sales volume of each product based on the daily inventory logs to identify the sales situation of each product in each store. Based on the built-in product categories, the sales performance of the same type of product is analyzed discretely to determine whether some products are selling worse than others. If it is determined that the sales performance of some products is worse than that of other products, these products are recorded as products to be evaluated. The sales performance of the corresponding products to be evaluated in each store is obtained, and the sales performance of the products to be evaluated is discretely judged to determine whether there are some stores whose sales performance is better than that of other stores. If it is determined that some stores are performing better than others, then the inventory of the products to be evaluated in the underperforming stores will be adjusted based on the sales performance of the top-performing stores. If it is determined that there are no stores where sales are better than other stores, then obtain the supplier information of the product to be evaluated, and read the corresponding supplier's purchase order based on the supplier information of the product to be evaluated. The corresponding supplier's purchase order is matched with the product to be evaluated. If the proportion of the product to be evaluated in the corresponding purchase order is greater than the built-in evaluation threshold, the supplier is deemed unqualified. If the proportion of the product to be evaluated in the corresponding purchase order is less than the built-in evaluation threshold, the supplier will be marked with a product based on the product to be evaluated, and information on suppliers with poor product markings will be obtained.

[0051] In this embodiment, daily inventory logs are analyzed to assess the sales volume of goods in each store. Discrete judgments are then made based on product categories to quickly identify products within the same store whose sales performance is inferior to similar products. This provides a clear direction for subsequent supplier selection optimization. Furthermore, by assessing the sales performance of the same product in different stores, regional factors are ruled out as the reason for the product's inferior sales performance. If it is determined that some stores are performing better than others, goods are reassigned to the stores with lower sales to avoid inventory backlog. If it is determined that no stores are performing better than others, the product itself is considered to have a problem. Therefore, the supplier of the product is assessed. If most products from the same supplier are marked as products to be evaluated, the supplier is considered to have quality issues and does not meet procurement standards, allowing for continuous supplier quality optimization and product quality improvement. If only a small number of products from the same supplier are marked as products to be evaluated, those products from that supplier are considered unqualified, and these products are marked to reduce future erroneous procurement.

[0052] For example, after each purchase, the sales performance of the purchased goods is assessed to evaluate the quality of the purchased goods, and in turn, to evaluate the quality of the corresponding supplier.

[0053] For example, if there is a store 1, and the sales of product a in store 1 are much lower than the sales volume of other similar products, then product a can be marked as a product to be evaluated. For example, if the sales volume of biscuit A is much lower than that of biscuit B, biscuit C, biscuit D, etc. at the same price, then biscuit A can be determined to be a product to be evaluated in store 1.

[0054] Next, statistical analysis is performed on the sales volume of biscuit A in stores 2, 3, ..., n. If only a small number of stores have similar sales volume of biscuit A as store 1, then the reason for the low sales volume of biscuit A in store 1 is determined to be due to regional characteristics. For example, users in the area where store 1 is located may not like biscuit A. Therefore, in this case, the inventory of biscuit A in store 1 can be redistributed to other stores with better sales.

[0055] If the sales volume of biscuit A in most stores is similar to that of store 1, then the reason for the low sales volume of biscuit A in store 1 is determined to be a quality problem with biscuit A (e.g., poor taste). In this case, by tracing the supplier of biscuit A, if the supplier purchases a total of biscuits A, sausages A, and fruits A, and most of these purchased items are marked as products awaiting evaluation (indicating quality problems), then the supplier's product quality is deemed poor, and the supplier is removed to maintain the quality of suppliers in the database. Conversely, if only a very small portion of the biscuits A, sausages A, and fruits A purchased by the supplier are marked as products awaiting evaluation, then the quality of individual products from that supplier is deemed poor. Therefore, that supplier's product category is marked to avoid incorrectly collecting data from that supplier in future data collection, thus improving the quality of purchased goods.

[0056] Compared with existing intelligent order scheduling methods, this invention realizes intelligent and optimized order scheduling, improves procurement efficiency and accuracy, and reduces procurement costs.

[0057] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. An intelligent order scheduling method, characterized in that, include: Obtain the order information to be scheduled, and perform statistical analysis on the product information of the goods to be purchased in the order information to obtain the goods to be purchased and their demand. Based on the order information, determine the reserved time slots for each order to purchase goods. Based on the category of goods to be purchased, statistically analyze the reserved time slots for the same category of goods to be purchased to obtain time gradient data, and obtain the gradient purchase quantity of each goods to be purchased under the corresponding time gradient. Obtain the production efficiency, current inventory, and delivery time of each supplier, and combine this with the time gradient data of each product to be purchased to determine the satisfaction level of each supplier for each product to be purchased, thereby obtaining the first supplier to be allocated that can fully satisfy the needs and the second supplier to be allocated that cannot fully satisfy the needs. For the first supplier to be allocated, the demand for the goods to be purchased is matched and calculated with the quantity and price comparison table of each supplier in the first supplier to be allocated, so as to obtain the purchase price of each supplier. The purchase prices are then compared and the first purchase plan with the lowest purchase price is selected. For the second supplier to be allocated, the supplier is combined and price is analyzed based on the corresponding supplier’s production efficiency, current inventory, delivery time, time gradient data of the goods to be purchased and the gradient purchase quantity under the corresponding time gradient to obtain the second purchase plan with the minimum purchase price. The prices of the first and second procurement plans are compared, and the procurement plan with the lowest procurement price is taken as the optimal allocation plan. A corresponding purchase order is generated and sent to the corresponding supplier.

2. The intelligent order scheduling method according to claim 1, characterized in that: The process of obtaining the order information to be scheduled and statistically analyzing the product information of the goods to be purchased in the order information to obtain the goods to be purchased and their demand quantities includes: Obtain the order information to be scheduled, and based on the store information and built-in product categories, classify the products and calculate the purchase volume of the orders submitted by the same store to obtain the category purchase volume of the same product category; Based on the store information, obtain the historical category purchase volume and corresponding inventory information of the corresponding store, judge the historical category purchase volume and corresponding inventory information, determine the dynamic balance relationship between the historical category purchase volume and the corresponding inventory information, and record it as the first balance relationship. Obtain the current store's inventory information, and compare the current store's inventory information with the category purchase quantity in the submitted order information to determine the dynamic balance relationship between the current category purchase quantity and the corresponding inventory information, and record it as the second balance relationship; The first balance relationship is compared with the second balance relationship. If the second balance relationship is greater than the first balance relationship, it is determined that the order information submitted by the store is abnormal.

3. The intelligent order scheduling method according to claim 2, characterized in that: The step of obtaining the order information to be scheduled, and statistically judging the product information of the goods to be purchased in the order information to obtain the goods to be purchased and their demand quantities, also includes: When it is determined that there are abnormalities in the order information submitted by the store in the order information to be dispatched, the daily inventory log of each product to be purchased in the order information of the corresponding product category is obtained, and the consumption status of each product to be purchased is determined based on the daily inventory log. Based on consumption data, forecast the consumption of goods for the next procurement cycle to obtain the expected consumption data, and determine the expected procurement quantity based on the expected consumption data and the current inventory level. Based on the product information of the goods to be purchased, the expected purchase quantity is compared with the purchase demand quantity in the order information. If the expected purchase quantity is greater than the purchase demand quantity, it is determined that the purchase demand quantity of the goods is normal. If the expected purchase quantity is less than the purchase demand quantity, the purchase demand quantity of the product is determined to be abnormal, and an alarm signal for abnormal purchase demand of the product is output.

4. The intelligent order scheduling method according to claim 1, characterized in that: The process of obtaining the production efficiency, current inventory, and delivery time of each supplier, and combining this with the time gradient data of each item to be purchased to determine the supplier's satisfaction level for each item, yields a first-tier supplier that can fully satisfy the demand and a second-tier supplier that cannot fully satisfy the demand. This includes: The production time is calculated by calculating the difference between the reserved gap time in each time gradient stage and the delivery time of each supplier. Based on production time and production efficiency, the quantity of goods that the supplier can produce under the corresponding time gradient is determined, and the production quantity is obtained. Compare the sum of production volume and the corresponding supplier's current inventory with the gradient purchase volume of the corresponding time gradient data; If the sum of the production volume and the current inventory of the corresponding supplier is greater than the gradient purchase volume of the corresponding time gradient data, then it is determined that the supplier can meet the corresponding purchase volume. Based on the suppliers' ability to meet the requirements of the purchased goods, suppliers who can meet the corresponding purchase volume of the purchased goods in each time gradient are designated as the first suppliers to be allocated, and suppliers who can only meet the purchase volume in some time gradients are designated as the second suppliers to be allocated.

5. The intelligent order scheduling method according to claim 4, characterized in that: Based on the suppliers' ability to meet the demand for goods, suppliers who can meet the corresponding purchase volume of the goods at each time gradient are designated as first-tier suppliers to be allocated, and suppliers who can only meet the purchase volume at some time gradients are designated as second-tier suppliers to be allocated, including: When it is determined that some suppliers cannot fully meet the tiered procurement quantities under each time gradient, the tiered procurement quantities of the time gradients that cannot be met are obtained and recorded as the tiered procurement quantities to be met. Based on time gradient data and unmet time gradients, obtain the gradient purchase quantity of the time gradient before the unmet time gradient time point, and record it as the met gradient purchase quantity. The difference between the tiered procurement quantity to be satisfied and the tiered procurement quantity already satisfied is calculated to obtain the compensation difference. Based on the compensation difference, the difference in the gradient procurement quantity of the time gradient located after the time gradient point that cannot be met is calculated to obtain the corrected comparison value. The corrected comparison value is compared with the sum of the supplier's production and current inventory under the corresponding time gradient until all time gradients have been compared. Suppliers are marked based on the time gradients that cannot be met, and a second supplier with a time gradient gap is obtained.

6. The intelligent order scheduling method according to claim 1, characterized in that: The second supplier to be allocated is selected by combining and analyzing suppliers based on their production efficiency, current inventory, delivery time, time-series data of the goods to be purchased, and the tiered purchase volume under the corresponding time-series, to obtain the second procurement plan with the lowest purchase price, including: Based on the supplier's production efficiency, current inventory, delivery time, and time gradient data of the goods to be purchased, determine the maximum gradient purchase quantity that each supplier in the second list of suppliers to be allocated can meet, and record it as the maximum supply quantity of the corresponding supplier. The difference between the maximum supply and the demand for the goods to be purchased is calculated to obtain the procurement difference data. Based on the procurement difference data and the maximum supply of each supplier, other suppliers other than the supplier are matched until the sum of the maximum supply of the combined suppliers is greater than the demand for the goods, thus obtaining the combined data. Select a set of combinations from the combined data, construct a cost function with conditional constraints, transform the cost function using a mixed integer nonlinear programming algorithm, and solve the transformed cost function using a business solver to obtain the optimal solution for the corresponding combination. The combination allocation result under the corresponding optimal solution is recorded as the combination allocation scheme, and the purchase price of the corresponding combination allocation result is recorded as the combination allocation price. Based on the combined data, the combined allocation prices under each combined case are compared with each other, and the combined allocation scheme and combined allocation price with the smallest purchase price in the combined data are selected. The selected result is recorded as the second purchase scheme.

7. The intelligent order scheduling method according to claim 1, characterized in that: Also includes: Obtain daily inventory logs for each product in each store, and determine the sales volume of each product based on the daily inventory logs to identify the sales situation of each product in each store. Based on the built-in product categories, the sales performance of the same type of product is analyzed discretely to determine whether some products are selling worse than others. If it is determined that the sales performance of some products is worse than that of other products, then these products are recorded as products to be evaluated. The sales performance of the corresponding products to be evaluated in each store is obtained, and the sales performance of the products to be evaluated is discretely judged to determine whether there are some stores whose sales performance is better than that of other stores. If it is determined that some stores are performing better than others, then the inventory of the products to be evaluated in the underperforming stores will be adjusted based on the sales performance of the top-performing stores. If it is determined that there are no stores where sales are better than other stores, then obtain the supplier information of the product to be evaluated, and read the corresponding supplier's purchase order based on the supplier information of the product to be evaluated. The purchase orders of the corresponding suppliers are matched with the products to be evaluated. If the proportion of the products to be evaluated in the corresponding purchase orders is greater than the built-in evaluation threshold, the supplier is deemed unqualified. If the proportion of the product to be evaluated in the corresponding purchase order is less than the built-in evaluation threshold, then the supplier is marked with a product label based on the product to be evaluated, and supplier information with poor product labels is obtained.

Images