Power material demand planning generation system with multi-objective resource optimization function

Abstract

The present invention relates to the field of computer automation, and in particular to the application of computer automation in the planning and storage management of materials in the power industry, and more specifically to a power material demand plan generation system with a multi-objective benefit optimization function, which can fully consider the particularity of material management in the power industry, and fully eliminate existing resources in advance when formulating demand plans, thereby improving the efficiency of demand plan formulation, eliminating and avoiding material backlogs. The power material demand plan generation system with a multi-objective benefit optimization function built on a network server of the present invention can be used across regions, thereby breaking down resource sharing barriers, opening up coordination channels, and realizing the rationalization and economic development of demand plans within the province.

Description

Technical Field

[0001] The present invention relates to the field of computer automation, and in particular to the application of computer automation in the planning and storage management of materials in the power industry, and more specifically to a power material demand plan generation system with a multi-objective profit and storage optimization function. Background Art

[0002] The procurement of materials in the power industry is characterized by large quantities, long cycles, and being easily affected by external environmental factors. Therefore, compared with the formulation of procurement strategies for other commodities, the procurement of materials in the power industry requires comprehensive consideration of aspects such as warehouse management and coordinated linkage allocation before procurement, and good demand planning, which is more difficult and challenging for demand planners and managers.

[0003] In the formulation of demand plans, the first step is to make use of the existing inventory, make rational use of the backlog inventory, and activate the backlog inventory. Improving the utilization rate of power materials is an important link before the formulation and distribution of the demand plan list. As an important component and management method of refined material management, material balance and utilization are of great significance to speeding up the effective turnover of inventory materials, reducing the risk of material damage / expiration caused by long-term backlog, promoting the rationalization and efficiency of the warehousing system, and improving resource utilization.

[0004] Since the power industry has many difficulties in balancing and utilizing materials, such as multiple specialties, multiple levels, difficulty in collaborative sharing, and lack of resource coordination, etc., in the process of formulating demand plans, fully considering existing resources and rationalizing and maximizing the utilization of materials is not only a hot topic for technical personnel, but also a difficulty and pain point in making demand plans. Summary of the invention

[0005] In order to improve the accuracy of material demand planning in the power industry and fully digest the existing resources before formulating the demand plan, the present invention discloses a power material demand plan generation system with a multi-objective benefit optimization function. The power material demand plan generation system with a multi-objective benefit optimization function is based on a network application server, which includes:

[0006] A material demand integration module, which is used to integrate demand items from different projects and group them according to material types from the demand items of each project to form a demand set for different materials;

[0007] The full material resource library includes all material resources available in the province, city, and county, including warehouse inventory resources, in-transit order resources, and agreement share resources;

[0008] Among them, warehouse inventory benefits come from the materials accumulated in various physical warehouses and the materials that are no longer used in the projects determined in the professional warehouses; in-transit order benefits come from the materials that have been in transit for a long time and have not been put into the warehouse due to design changes and other reasons; the agreement share benefits come from a certain proportion of the agreement shares reserved in advance at the provincial company level and the shares of materials that have been signed but have not been fulfilled for a certain period of time due to various reasons;

[0009] A benefit resource strategy generation module, wherein the benefit resource strategy generation module comprises a benefit resource data cache submodule, a benefit resource algorithm submodule, and a benefit resource solution result cache submodule. The benefit resource data cache submodule is used to temporarily store a benefit resource data set for benefit algorithm calculation retrieved from the full material resource library according to the benefit resource strategy. The benefit resource algorithm submodule is used to calculate and generate a benefit optimal solution. The benefit solution result cache submodule is used to temporarily store each benefit solution in the optimal solution set calculated in the benefit algorithm submodule.

[0010] The data cache submodule of the resource library is connected with the full material resource library through a data interface;

[0011] The benefit strategy is to benefit the users in a step-by-step manner. First, benefit the users at the county level, and collect the needs within the county into the county benefit resource data set for benefit. Then, benefit the users at the city level, and collect the unsatisfied needs into the city benefit resource data set for benefit. Then, benefit the users at the provincial level, and collect the needs that are still unsatisfied after two benefits into the provincial benefit resource data set for benefit. Finally, the three-level benefit solution result cache is placed in the benefit solution result cache submodule.

[0012] A benefit solution optimization module, which is connected to the benefit strategy generation module through a data interface, and extracts each benefit solution in the benefit solution result cache submodule through the data interface, and calculates according to the optimization conditions, eliminates inferior solutions, and forms an optimized benefit solution;

[0013] The demand plan generation module amends and returns the demand items of each project according to the optimized resource utilization plan obtained in the resource utilization plan optimization module, and generates a scheduling list.

[0014] The power material demand plan generation system with multi-objective resource optimization function disclosed in the present invention can match the demand items of different projects with resource available at different levels, in different regions and from different sources, thereby maximizing the consumption of existing material resources through "one to one", "one to many", "many to one" and "many to many" methods, and the demand plan formed on this basis is more reasonable.

[0015] Further preferably, a manual verification module is also included. After the demand plan optimization plan is generated, the demand items and scheduling lists of each project are revised and returned, and are further manually reviewed in the manual verification module. After confirmation, a formal scheduling list and return and modification items are generated.

[0016] After the entire demand plan generation step is completed, the resources not involved in the scheduling list are released and rolled back to the full material resource library by the resource library data submodule in the resource library strategy generation module.

[0017] The intervention of manual review can improve the maneuverability and flexibility of the strategy. For some special situations, the design of the manual verification module can ensure timely response to these situations.

[0018] Further preferably, the algorithm steps adopted by the Liku algorithm submodule are as follows:

[0019] S1. Output the maximum benefit solution set. Assuming there are n demand sets {prjitem1, prjitem2, ... prjitem n} and m benefit resource sets {mltem1, mltem2, ... mltem m}, outputting the maximum benefit solution set includes the following steps:

[0020] Step 1: Input n demand items and m available resource items of a certain material;

[0021] Step 2: For each demand item and the corresponding profitable resource item, calculate the profitable quantity DQ in the intersection item. The DQ quantity depends on whether the material supply item mltem i meets the special requirements of the corresponding demand. If it does, it is the remaining profitable quantity of the profitable resource item. If it does not meet the demand restrictions, it is marked as 0.

[0022] Step 3: For the first demand prjitem1, search for matches in the corresponding m stocks one by one. If the DQ quantity of the corresponding item is greater than 0, mark the matching profit quantity x i1 If the DQ quantity is 0, it means that the profitable resource entry has no profitable resource, and it is marked as x i1 is 0, that is, x i1 =min(DQ1,total quantity of inventory item i) or 0;

[0023] Step 4: When Σx i1 = the number of requirement item prjitem1, the matching of this requirement item is completed, and the next requirement item is continued to execute step 2 until all the loop matching is completed;

[0024] Step 5: Add the n maximization benefit results after each set of demand matching to the solution set, which can support the output of the optimal demand solution according to other objectives;

[0025] S2. Generate the optimal benefit plan: Use the known benefit plan set as the initial data source, and solve the optimal matching results one by one for different target levels of the balanced benefit business.

[0026] For example, the n benefit pool schemes that have achieved the maximum benefit pool are selected in turn according to the target level.

[0027] There are m goals, arranged in order of importance as follows:

[0028] f1(x),f2(x),f3(x)......f m (x)

[0029] According to the actual needs of balancing the library, selective settings:

[0030] ①Important goal f1(x): the difference between the demand items and the available resource items is minimal;

[0031] ② Secondary goal f2(x): Minimize the number of returned and modified items (the total number of returned items and the number of returned and modified items);

[0032] A1: First, find the optimal solution for the important goal, that is, the minimum difference between the demand items and the available resource items, and find the set R0 of all optimal solutions;

[0033] A2: Find the optimal solution of the secondary goal within R0, and record the set of optimal solutions as R1, until the optimal solution of the mth goal is obtained. The model is as follows:

[0034]

[0035]

[0036] …

[0037]

[0038] This method calculates R0, R1, ... m-1 It must be non-empty, otherwise it cannot be performed.

[0039] By using the algorithm disclosed in the present invention to perform strategic operations, it is possible to reduce the refund and change items on the basis of maximizing the elimination of existing resources, so that the impact of the benefit warehouse process on the declaration of project demand is minimized. Under this optimal solution algorithm system, differential benefit warehouses are allowed, so that inventory can be reduced to the greatest extent, and the number of refund and change items can be minimized, achieving the optimal match between the demand set and the benefit warehouse resource set, significantly saving the workload of material planners, while avoiding unnecessary secondary reporting of plans and shortening the balance benefit warehouse circulation cycle, which has more obvious advantages.

[0040] Further preferably, in the present invention, for different types of materials, the benefit-stock algorithm submodule performs benefit-stock scheme calculation according to the following steps:

[0041] The first step is to determine the type of materials. If they are ordinary materials that can be divided into cases, they are transferred to the first type of materials. If they are measured by length or weight, such as cables or installation materials, they are transferred to the second type of materials.

[0042] Step 2:

[0043] If it is a first type of material, the relationship between the total number of resources that can be used to benefit the material and the minimum required quantity is further determined. If the total number of resources that can be used to benefit the material is greater than the minimum required quantity, benefit is obtained according to K1; if the total number of resources that can be used to benefit the material is less than the minimum required quantity, benefit is obtained according to K2.

[0044] If it is a second type of material, further determine whether the material is a demand item with a section length requirement. If it is a demand item with a section length requirement, then benefit the inventory according to K3; if it is a demand item without a section length requirement, then benefit the inventory according to K4;

[0045] The K1, K2, K3, and K4 mentioned are as follows:

[0046] K1: Arrange and combine n demands that are less than the total number of available resources to form a demand plan combination of {m1, m2, m3, m4, ...};

[0047] Arrange and combine all x items of the profit pool resources to form b combinations of profit pool items {y1, y2, y3, y4, ...};

[0048] (a) A certain demand quantity is matched with a certain resource quantity in a profit pool in a perfect match; m1=y1, eliminating the corresponding materials and demand;

[0049] (b) Prioritize matching when the quantity of multiple demands is exactly the same as the quantity of a certain profitable resource; m2+m3=y2, eliminating the corresponding materials and demands;

[0050] If there are multiple combinations with the same difference, they will be matched in the order of the total demand from large to small and the number of demand items from small to large;

[0051] for example:

[0052] m4+m5+m6=y3

[0053] m7+m8=y4

[0054] m9+m10=y5,

[0055] First, compare the size of y and prioritize eliminating the ones with larger y values; then compare the number of m entries and prioritize eliminating the ones with fewer m entries;

[0056] K2: Select the demand with the smallest difference with the total quantity of available resources for matching. If the difference between the quantity of demand and available resources of multiple demand plans is the same, a demand plan will be randomly selected for matching. In other words, try to use one demand item to benefit the resource first, and then allocate it when there is still some left.

[0057] K3: If the length of any profitable resource is less than the required segment length, no profit will be made; if there is a profitable resource with a length greater than the required segment length, all profitable resources with a length greater than the required segment length will be selected for permutation and combination, and the group with a total length greater than the total required length and the smallest difference will be profited.

[0058] For example, the available resources are y1, y2, y3, etc., and the demand is x1, x2, x3, etc.

[0059] Compare x1 with all y values. If y is less than x, it is disadvantageous to store. If y1>x1+x2, y2>x3+x1+x2, compare y1-x1-x2 and y2-x1-x2-x3. If they are greater than x, the one with the smaller difference, i.e. y2, will be eliminated first.

[0060] K4: If the total length of the materials in the warehouse is less than the minimum demand, the minimum demand is selected for matching; if the total length of the materials in the warehouse is greater than the minimum demand, the demand items and the length of the materials in the warehouse are arranged and combined respectively, and the one with the smallest difference is selected for matching.

[0061] For example, the available resources are y1, y2, y3, etc., and the demand is x1, x2, x3, etc.;

[0062] Compare x1 and y. If both y are less than x, select the smallest x. If y1>x1+x2, y2>x3+x1+x2, compare y1-x1-x2 and y2-x1-x2-x3. If they are greater than each other, give priority to eliminating the one with the smaller difference, i.e. y2.

[0063] Different warehouse utilization operation strategies are formed for different materials, so as to achieve targeted and refined warehouse utilization and further improve the warehouse utilization effect.

[0064] Further preferably, different priority levels of the benefit resources within the same benefit range are considered in the benefit algorithm submodule operation, and the operation result is generated according to the following priority order:

[0065] Warehouse inventory is the highest priority resource, in-transit order resources are the second priority resource, and agreement share resources are the lowest priority resource.

[0066] It should be noted that within the scope of county-level profit-making, since there are no agreed share profit-making resources, only warehouse inventory profit-making resources and in-transit order profit-making resources are considered, and priority is given to profit-making among warehouse inventory profit-making resources. When the warehouse inventory profit-making resources are completely reduced or cannot meet the profit-making needs, the in-transit order profit-making resources are added to reduce the profit-making.

[0067] Further preferably, different priority levels of the benefit resources within the same benefit range are considered in the benefit algorithm submodule operation, and the operation result is generated according to the following priority order:

[0068] For warehouse inventory that can be used for storage, the distance between the storage resource location and the demand destination> the storage difference> the storage age of the stored resources> the number of factories involved in the storage plan (the number of county-level factories is constant at one);

[0069] Specifically, when further considering specific influencing factors, the distance is preferably short, the difference in the utilization of resources is preferably small, the age of the utilized resources is preferably long, and the number of factories involved is preferably small.

[0070] For in-transit orders, you can use the warehouse resources. The number of factories involved in the warehouse plan > the warehouse difference > the time of in-transit orders.

[0071] Specifically, when further considering specific influencing factors, the number of factories involved is preferably small, the profit and inventory difference is preferably a small difference, and the order time in transit is preferably a longer time in transit.

[0072] For the agreed share of profit-making resources, the contract signing time > contract execution progress > profit difference;

[0073] Specifically, when further considering specific influencing factors, the contract signing time is preferably an agreement with a longer signing time, the contract execution progress is preferably an agreement with a shorter execution progress, and the profit and inventory difference is preferably a small difference.

[0074] As a preferred technical solution, a maximum matching threshold of available inventory resources is preset in the inventory optimization module. When the number of warehouses where available inventory resources matching a certain material in the inventory plan are located exceeds the maximum matching threshold, the inventory plan is an inferior solution.

[0075] More preferably, the threshold is 3.

[0076] In a preferred technical solution, an economic value judgment condition is set in the benefit plan optimization module. When a benefit plan is generated, the inventory cost reduction of the benefit plan and the transportation cost generated by the benefit transfer are compared:

[0077] If the inventory cost reduced by the stock-sharing plan is greater than the transportation cost incurred by the stock-sharing transfer, then the stock-sharing plan is the optimal solution;

[0078] If the inventory cost reduced by the inventory transfer plan is less than the transportation cost incurred by the inventory transfer, then the inventory transfer plan is an inferior solution.

[0079] The purpose of doing this is that since warehouses are relatively dispersed, the transportation costs generated by the distance between the source warehouses and the destination warehouses during the cross-regional and cross-level warehousing process may be greater than the cost of continuing inventory. Such warehousing has no economic value and is therefore abandoned.

[0080] Further preferably, a minimum benefit unit threshold for benefitable resources is also set in the benefit plan optimization module. When the quantity of a certain material in the benefit data submodule is less than the minimum benefit unit threshold, the warehouse will not participate in the benefit.

[0081] As a preferred technical solution, the power material demand plan generation system with multi-objective resource optimization function is compatible with the ERP system. The full material resource library is a resource library formed by resources that meet the resource requirements in the ERP system. The resource data submodule captures the resource information within the resource range from the ERP system through the data interface and temporarily stores it in the resource data submodule.

[0082] As a preferred technical solution, the resource tags of the beneficial stock in the full material resource library have a mapping relationship. When the beneficial stock data cache submodule retrieves the beneficial stock resource data set used for beneficial stock algorithm calculation from the full material resource library, the old materials and replaceable material resources that have a mapping relationship with the original materials in demand are retrieved and stored in the beneficial stock data cache submodule.

[0083] The mapping relationship here includes two types of mapping, one is the mapping between new and old materials, and the other is the mapping between replaceable materials. Due to the rapid development of the power material industry, many specifications and product names change rapidly. Through mapping labels, these materials can be associated with each other, and old materials can be stored. At the same time, for materials that can achieve the same function and have a replacement relationship, they can be associated through replaceable label mapping, so that some of the materials that can be used in the warehouse can be used to achieve the purpose of using the warehouse for the purpose of realizing the function.

[0084] On this basis, further, in the calculation of the benefit algorithm sub-module, different benefit priority levels of the benefitable resources within the same benefit range are considered, and benefit is performed according to the following priority: complete benefit of old materials > complete benefit of original materials > complete benefit of replaceable materials > partial benefit of old materials > partial benefit of original materials.

[0085] The demand planning intelligent auxiliary system disclosed in the present invention can fully consider the particularity of material management in the power industry, and fully eliminate existing resources in advance when formulating demand plans, thereby improving the efficiency of demand planning schemes and eliminating and avoiding material backlogs. The power material demand plan generation system with multi-objective profit and storage optimization function built on the basis of a network server of the present invention can be used across regions, thereby breaking down resource sharing barriers, opening up coordination channels, and realizing the rationalization and economic development of demand plans within the province. DETAILED DESCRIPTION

[0086] In order to better understand the present invention, the present invention is further described below in conjunction with specific embodiments.

[0087] Example 1

[0088] The invention discloses a system for generating a demand plan for electric power materials with a multi-objective benefit optimization function. The system for generating a demand plan for electric power materials with a multi-objective benefit optimization function is based on a network application server. Meanwhile, the system for generating a demand plan for electric power materials with a multi-objective benefit optimization function is compatible with an ERP system. The PC terminals of various projects and management offices are connected to the system for generating a demand plan for electric power materials with a multi-objective benefit optimization function disclosed in the invention through a data interface.

[0089] The network application server of the power material demand plan generation system with multi-objective resource optimization function includes:

[0090] A material demand integration module, which is used to integrate demand items from different projects and group them according to material types from the demand items of each project to form a demand set for different materials;

[0091] The full material resource library includes all material resources available in the province, city, and county, including warehouse inventory resources, in-transit order resources, and agreement share resources;

[0092] Among them, warehouse inventory benefits come from the materials accumulated in various physical warehouses and the materials that are no longer used in the projects determined in the professional warehouses; in-transit order benefits come from the materials that have been in transit for a long time and have not been put into the warehouse due to design changes and other reasons; the agreement share benefits come from a certain proportion of the agreement shares reserved in advance at the provincial company level and the shares of materials that have been signed but have not been fulfilled for a certain period of time due to various reasons;

[0093] In this embodiment, since the PC terminals of each project are connected to the power material demand plan generation system with multi-objective resource optimization function disclosed in the present invention through a data interface, the required data can be directly retrieved from each PC terminal through the data interface.

[0094] A benefit resource strategy generation module, wherein the benefit resource strategy generation module comprises a benefit resource data cache submodule, a benefit resource algorithm submodule, and a benefit resource solution result cache submodule. The benefit resource data cache submodule is used to temporarily store a benefit resource data set for benefit algorithm calculation retrieved from the full material resource library according to the benefit resource strategy. The benefit resource algorithm submodule is used to calculate and generate a benefit optimal solution. The benefit solution result cache submodule is used to temporarily store each benefit solution in the optimal solution set calculated in the benefit algorithm submodule.

[0095] The data cache submodule of the resource library is connected with the full material resource library through a data interface;

[0096] The benefit strategy is to benefit the users in a step-by-step manner. First, benefit the users at the county level, and collect the needs within the county into the county benefit resource data set for benefit. Then, benefit the users at the city level, and collect the unsatisfied needs into the city benefit resource data set for benefit. Then, benefit the users at the provincial level, and collect the needs that are still unsatisfied after two benefits into the provincial benefit resource data set for benefit. Finally, the three-level benefit solution result cache is placed in the benefit solution result cache submodule.

[0097] A benefit solution optimization module, which is connected to the benefit strategy generation module through a data interface, and extracts each benefit solution in the benefit solution result cache submodule through the data interface, and calculates according to the optimization conditions, eliminates inferior solutions, and forms an optimized benefit solution;

[0098] The demand plan generation module amends and returns the demand items of each project according to the optimized resource utilization plan obtained in the resource utilization plan optimization module, and generates a scheduling list.

[0099] Below we further illustrate the algorithm operation mode disclosed in the present invention through an example.

[0100] For example, after integration by the material demand integration module, taking a certain material transformer as an example, the transformer demand set of A County, X City is (4, 7, 10, 12, 17, 23), the transformer demand set of B County, X City is (8, 14), and the demand set of M County, Y City is (9, 12, 15).

[0101] According to the three-level resource library hierarchy of county level, city level and provincial level, firstly, for the transformer demand set of County A in City X, which is (4, 7, 10, 12, 17, 23), the resource library data cache submodule retrieves the resource library resources of County A from the full material resource library through the data interface, and temporarily stores the retrieved resources XA warehouse inventory (10, 16) and XA in-transit order resources (7) in the resource library data cache submodule.

[0102] Then, the profit-sharing algorithm submodule generates the profit-sharing optimal solution. Specifically, in this embodiment, the transformer demand set of County A in City X is (4, 7, 10, 12, 17, 23), the profit-sharing resources of the XA warehouse inventory are (7, 10, 16), and the profit-sharing resources of the XA in-transit orders are (7).

[0103] According to the preferred logic of benefiting from the stockpiles disclosed in the present invention, transformers are common materials that can be separated by pieces. Therefore, benefiting from the stockpiles is carried out according to the conditions. When the total quantity of materials is greater than the minimum demand quantity: all x items of benefiting from the stockpiles are arranged and combined to form b combinations of benefiting from the stockpiles of {y1, y2, y3, y4, ...}; (a) priority matching when a certain demand quantity is exactly the same as a certain benefiting from the stockpiles; m1 = y1, eliminating the corresponding materials and demands; (b) priority matching when multiple demand quantities are exactly the same as a certain benefiting from the stockpiles; when the total quantity of materials is greater than the minimum demand quantity, the demand with the smallest difference with the total quantity of benefiting from the stockpiles is selected for matching. If the difference between the demand quantities of multiple demand plans and the benefiting from the stockpiles is the same, a demand plan is randomly selected for matching. In other words, try to benefit from one demand item first, and then distribute it when there is still a surplus.

[0104] In this embodiment, according to the above rules, for the demand of County A in City X, the profit library includes the following steps:

[0105] In the first round of benefit sharing, the demand of County A, City X is 7, which is the same as the 7 in-transit benefit sharing resources of XA and the 7 in the warehouse benefit sharing resources of XA. According to the benefit sharing priority, 7 in the warehouse benefit sharing resources of XA in-transit orders is shared with the demand of County A, City X. At the same time, the demand of County A, City X is 10, which is the same as the 10 in the warehouse stock of XA. Therefore, 10 in the warehouse stock of XA is shared with the demand of County A, City X, 10.

[0106] After this round, the remaining demand in County A, City X is (4, 12, 17, 23). Meanwhile, the remaining resources in the first profit pool are XA warehouse inventory (16), and XA's in-transit profit pool resources are (7).

[0107] Then, the second round of profit sharing is carried out, and the remaining demands (4, 12, 17, 23) of City X are arranged into 4, 12, 16 (4+12), 17, 21 (4+17), 27 (4+23), 29 (12+17), 33 (4+12+17), 35 (12+23), 39 (4+12+23), 40 (17+23), 44 (4+17+23), and 52 (12+17+23). The cases where the remaining demand combination is consistent with the single inventory in the XA warehouse inventory are screened, and it is found that the remaining demand combination 16 of County A in City X is consistent with the inventory 16 of the XA warehouse, so a direct match is made, that is, the inventory 16 of the XA warehouse is allocated to the demand combination 16 of County A in City X, that is, the combination of demand 4 and demand 12.

[0108] After this round of profit sharing, the remaining demand in County A, City X is (17, 23), the remaining inventory in warehouse XA is 0, and the available profit sharing resources in transit in warehouse XA are (7).

[0109] At this time, there are no physical inventory resources available, so we further use the local in-transit resources at the county level. Based on the method of minimizing the difference, 7 of XA's in-transit resources will be given to the demand 17.

[0110] After this round of profit sharing is completed, the remaining demand of A in City X is (10, 23), the remaining inventory in XA warehouse is 0, and the available profit sharing resources in transit are also 0.

[0111] At this time, since there is still demand for transformers in County B, which is also City X, the remaining demand (10, 23) of City X A will not be benefited at the second level (city level), but County B, which is also at the county level, will continue to benefit at the county level.

[0112] The Liku data cache submodule retrieves the available Liku resources in County B from the full material resource library through the data interface, and temporarily stores the retrieved resource XB warehouse inventory (3, 8, 9, 10, 14) in the Liku data cache submodule.

[0113] Then, the profit-pool algorithm submodule generates the profit-pool optimal solution. Specifically, in this embodiment, the transformer demand set of County B in City X is (8, 14), and the inventory of the XB warehouse is (3, 8, 9, 10, 14).

[0114] After the same county-level benefit storage, all the demand of County B in City X is met and is no longer incorporated into the demand of City X. The remaining inventory in the XB warehouse is (3, 9, 10).

[0115] At this time, since all the county-level benefits in City X have been completed, the transformer demand of City X is integrated to form the demand of City X (10, 23). At the same time, the benefit data cache submodule retrieves the available benefits of City X from the full material resource library through the data interface and temporarily stores them in the benefit data cache submodule. At this time, since the available benefits of transformers in City X only exist in County A and County B, no new available benefits are introduced, and the remaining resources in County B, the remaining inventory of XB warehouse, are (3, 9, 10).

[0116] The profit-stock algorithm submodule generates the optimal solution for profit-stock. Specifically, in this embodiment, the demand of city X is arranged into 10, 23, and 33 (10+23), and the remaining inventory of warehouse XB is arranged into 3, 9, 10, 12 (3+9), 13 (3+10), and 19 (9+10). 10 of the profit-stock resources in warehouse XB are matched to the demand 10.

[0117] After this round of profit sharing, the remaining demand in City X is (23), and the remaining inventory in warehouse XB is (3, 9).

[0118] Then add up the remaining stocks in warehouse XB and give all of them to the remaining demand of city X. The final remaining demand of city X is 11.

[0119] At this time, since there is still transformer demand for the M County project in J Province, the remaining demand (11) of X City will not be allocated at the third level (provincial level), but the demand of M County in J Province will continue to be allocated.

[0120] The method is the same as that of City X.

[0121] During the first round of inventory clearing, since no situation was found where the demand was completely consistent with a single piece of inventory, a second round of inventory clearing was carried out.

[0122] The demands of city Y are arranged in different combinations to form 9, 12, 15, 21 (9+12), 24 (9+15), 27 (12+15), and 36 (9+12+15). Again, no situation is found that is completely consistent with the single inventory. Therefore, the third round of inventory is carried out.

[0123] Combine the YM warehouse to form 5, 14, 19 (5+14), match them according to the principle of minimum difference, match 14 in the combination to 12 in the demand, and mark the remaining inventory as 2. After this round of profit, the demand of M County in Y City is (9, 15), and the inventory of YM warehouse is (5, 2).

[0124] Since there is no other demand in Y City within the city level, the demand in Y City is (9, 15) and the inventory in YM warehouse is (5, 2).

[0125] The total inventory of warehouse YM is 7. According to the rule of minimum difference, the demand for warehouse Y is 9. At this time, the demand for warehouse Y is (2, 15).

[0126] At this time, since the city-level resource warehouses in Province J have been completed, the transformer demands in Province J are integrated to form the demand of Province J (11, 2, 15). At the same time, the resource warehouse data cache submodule retrieves the resource warehouse resources of Province J from the full material resource library through the data interface and temporarily stores them in the resource warehouse data cache submodule. At this time, the resource ZN warehouse (6) of the N warehouse in City Z is introduced.

[0127] At this time, the remaining demand is (11, 2, 15). According to plan K2, ZN warehouse (6) selects the demand with the smallest difference for matching. First, inventory 6 is allocated to demand 2, and the remaining inventory 4 is allocated to demand 15. At this time, there are no more available resources, and the final remaining demand is (11, 11).

[0128] After the inventory utilization strategy generation module, only 23 of the demands (4, 7, 10, 12, 17, 23) of County A in City X are not fully utilized, all the demands (8, 14) of County B in City X are fully utilized, and only 15 of the demands (9, 12, 15) of City Y are not fully utilized, and the remaining 11, all other demands are met by utilization. At the same time, in the process of utilization, the material resources of XA warehouse inventory (7, 10, 16), XB warehouse inventory (3, 8, 9, 10, 14), YM warehouse inventory (5, 14), XA in-transit order resources (7), and ZN warehouse inventory (6) are all reduced.

[0129] The benefit plan result cache submodule temporarily stores the benefit plan formed in the benefit calculation, and further compares the benefit plan to reduce inventory costs and transportation costs generated by benefit transfer according to the maximum matching threshold 3 of the benefit resources preset in the benefit plan optimization module and the economic value judgment conditions set in the benefit plan optimization module. Finally, the benefit plans formed in this embodiment are all non-inferior solutions.

[0130] Finally, through the demand plan generation module, according to the optimal benefit plan obtained in the benefit plan optimization module, the demand items of each project are corrected and returned, and a scheduling list is generated. According to the benefit results in this embodiment, the transformer demand plan for County A in City X is (11), and the transformer demand plan for County M in City Y is (11). At the same time, other benefit plans are automatically generated in the form of a scheduling list.

[0131] These scheduling lists are automatically fed back into the ERP system through the data interface with the ERP system.

[0132] Example 2

[0133] In order to illustrate the selection method of different profit influencing factors, in this embodiment, we use the following examples for explanation:

[0134] When the demand in a certain county is 4, 7, and 8, and there are three warehouse inventory resources 6, 8 (warehouse age 1 year), and 8 (warehouse age 2 years) in the county's warehouse, the warehouse inventory resource 8 with a warehouse age of 2 years will be preferentially allocated to the demand quantity 8.

[0135] When the demands of a county are 5, 8, and 8, and there are two available resources for in-transit orders in the county, namely 5 (in-transit time is 4 months) and 5 (in-transit time is 3 months), the resource with in-transit time of 4 months will be prioritized to the demand of 5.

[0136] Example 3

[0137] In order to illustrate the selection method of the Keliku resources with the mapping relationship between new and old materials, in this embodiment, we use the following example to illustrate:

[0138] If a county demands 14 and 25, and there are four warehouse inventory resources 3 (old materials with mapping relationship) in the county warehouse, 5, 6, and 9, the inventory is arranged into 3, 5, 6, 8 (3+5), 9, 9 (3+6), 11 (5+6), 12 (3+9), 14 (5+9), 14 (3+5+6), 15 (6+9), 17 (3+5+9), 18 (3+6+9), and 20 (5+6+9). The combination demand 14 is consistent with the inventory quantity, which are 3, 5, 6 and 5, 9 respectively. Among them, 3 is the old model of the material, so 3, 5, 6 will be preferred for inventory utilization.

[0139] If a county has demands of 2, 6, and 9, and there are three warehouse inventory resources in the county warehouse that can be used: 8 (old materials with a mapping relationship), 11 (replaceable materials), and 13 (original materials). If the demands are arranged and combined into 2, 6, 8, 9, 11, 15, and 17, the best matching combination 8 (2 and 6) or combination 11 (2 and 9) is selected for matching. Since 8 is an old material with a mapping relationship, it has the highest priority, so the old material is first eliminated according to the combination of 2 and 6. Then the remaining demand 9 is in the inventory 11 and 13. According to the priority order, since neither can be fully used, the original material is selected for partial use, and 13 is selected for use, and the remaining inventory is 11 (replaceable materials) and 4 (original materials). If the demand here is not 9, but 11, then because the full use of replaceable materials takes precedence over the partial use of original materials, 11 is selected first, and then the remaining inventory is 13 (original materials).

[0140] The above is a specific implementation of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications are also considered to be within the protection scope of the present invention.

Claims

1. A power material demand plan generation system with multi-objective resource optimization function, characterized in that: The power material demand plan generation system with multi-objective benefit storage optimization function is based on a network application server, which includes: A material demand integration module, which is used to integrate demand items from different projects and group them according to material types from the demand items of each project to form a demand set for different materials; The full material resource library includes all material resources available in the province, city, and county, including warehouse inventory resources, in-transit order resources, and agreement share resources; Among them, warehouse inventory benefits come from the materials accumulated in various physical warehouses and the materials that are no longer used in the projects determined in the professional warehouses; in-transit order benefits come from the materials that have been in transit for a long time and have not been put into the warehouse due to design changes; the agreement share benefits come from a certain proportion of the agreement shares reserved in advance at the provincial company level and the shares of materials that have been signed but have not been fulfilled for a certain period of time due to various reasons; A benefit resource strategy generation module, wherein the benefit resource strategy generation module comprises a benefit resource data cache submodule, a benefit resource algorithm submodule, and a benefit resource solution result cache submodule. The benefit resource data cache submodule is used to temporarily store a benefit resource data set for benefit algorithm calculation retrieved from the full material resource library according to the benefit resource strategy. The benefit resource algorithm submodule is used to calculate and generate a benefit optimal solution. The benefit solution result cache submodule is used to temporarily store each benefit solution in the optimal solution set calculated in the benefit algorithm submodule. The resource library data cache submodule is connected to the full resource library through a data interface; A benefit solution optimization module, which is connected to the benefit strategy generation module through a data interface, and extracts each benefit solution in the benefit solution result cache submodule through the data interface, and calculates according to the optimization conditions, eliminates inferior solutions, and forms an optimized benefit solution; The demand plan generation module amends and returns the demand items of each project according to the optimized resource utilization plan obtained in the resource utilization plan optimization module, and generates a scheduling list; For different types of materials, the profit algorithm submodule performs profit plan calculation according to the following steps: The first step is to determine the type of materials. If they are ordinary materials that can be separated by piece, they are transferred to the first type of materials. If they are cables or installation materials measured by length or weight, they are transferred to the second type of materials. In the second step, if it is the first type of material, the relationship between the total number of resources that can be used for the material and the minimum required quantity is further determined. If the total number of resources that can be used for the material is greater than the minimum required quantity, the material is used for the material according to K1; if the total number of resources that can be used for the material is less than the minimum required quantity, the material is used for the material according to K2. If it is a second type of material, further determine whether the material is a demand item with a section length requirement. If it is a demand item with a section length requirement, then benefit the inventory according to K3; if it is a demand item without a section length requirement, then benefit the inventory according to K4; The K1, K2, K3, and K4 mentioned are as follows: K1: Arrange and combine n demands that are less than the total number of available resources to form a demand plan combination of {m1, m2, m3, m4, ...}; Arrange and combine all x items of the profit pool resources to form b combinations of profit pool items {y1, y2, y3, y4, ...}; If the quantity of a certain demand is consistent with the quantity of a certain profitable resource, it will be matched first; m1=y1, eliminating the corresponding materials and demand; If the quantity of multiple demands is consistent with the quantity of a certain profitable stockpile resource, priority matching will be given; m2+m3=y2, eliminating the corresponding materials and demands; If there are multiple combinations with the same difference, they will be matched in the order of the total demand from large to small and the number of demand items from small to large; K2: Select the demand with the smallest difference with the total quantity of available resources for matching. If the difference between the quantity of demand and available resources of multiple demand plans is the same, a demand plan is randomly selected for matching. First, one demand item is used to benefit the resource. If there is still a surplus, it is allocated. K3: If the length of any profitable resource is less than the required segment length, no profit will be made; if there is a profitable resource with a length greater than the required segment length, all profitable resources with a length greater than the required segment length will be selected for permutation and combination, and the group with the combined length greater than the total required length and the smallest difference will be profited; K4: If the total length of the materials in the warehouse is less than the minimum demand, the minimum demand is selected for matching; if the total length of the materials in the warehouse is greater than the minimum demand, the demand items and the length of the materials in the warehouse are arranged and combined respectively, and the one with the smallest difference is selected for matching.

2. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: It also includes a manual verification module. After the demand plan optimization plan is generated, the demand items and scheduling lists of each project are revised and returned for further manual review in the manual verification module. After confirmation, the formal scheduling list and return and modification items are generated.

3. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: The algorithm steps adopted by the Liku algorithm submodule are as follows: S1. Output the maximum benefit solution set. Assuming there are n demand sets {prjitem 1, prjitem 2, ... prjitemn} and m benefit resource sets {mltem 1, mltem 2, ... mltem m}, outputting the maximum benefit solution set includes the following steps: Step 1: Input n demand items and m available resource items of a certain material; Step 2: For each demand item and the corresponding profitable resource item, calculate the profitable quantity DQ in the intersection item. The DQ quantity depends on whether the material supply item mltem i meets the special requirements of the corresponding demand. If it does, it is the remaining profitable quantity of the profitable resource item. If it does not meet the demand restrictions, it is marked as 0. Step 3: For the first demand prjitem 1, search and match in the corresponding m stocks in turn. If the DQ quantity of the corresponding item is greater than 0, mark the matching profit quantity xi1. If the DQ quantity is 0, it means that there is no profit resource for this profit resource item, and mark xi1 as 0, that is, xi1 = min (DQ1, the total quantity of inventory item i) or 0; Step 4: When Σxi1 = the number of requirement items prjitem 1, the matching of the requirement item is completed, and the next requirement item is continued to execute step 2 until all the loop matching is completed; Step 5: Add the n maximization benefit results after each set of demand matching to the solution set, and support the output of the optimal demand solution based on other objectives; S2. Generate the optimal benefit plan: Take the known benefit plan set as the initial data source, and solve the optimal matching results for different target levels of the balanced benefit business one by one, including the following steps: Step 1: For the above n maximum benefit pool results that have achieved maximum benefit pool, select them in turn according to the target level. There are m targets, which are arranged in the order of importance as follows: f1(x),f2(x),f3(x)......f m (x) Step 2: Compare with the actual demand of the balance library, and selectively set: important goal f1(x): the difference between the demand items and the available library resource items is the smallest; secondary goal f2(x): the number of refund and change items is the smallest; First, find the optimal solution for the important goal, that is, the minimum difference between the demand items and the available resource items, and find the set R0 of all optimal solutions; Find the optimal solution of the secondary goal in R0, and record the set of optimal solutions as R1, until the optimal solution of the mth goal is obtained. The model is as follows: ； In this model, R0, R1, ... m-1 It must be non-empty, otherwise it cannot be performed.

4. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: In the calculation of the benefit algorithm submodule, different benefit priority levels of the benefit resources in the same benefit resource data set are considered, and the calculation results are generated according to the following priority order: Warehouse inventory is the highest priority resource, in-transit order is the second priority resource, and agreement share is the lowest priority resource. For warehouse inventory that can be used for storage, the distance between the storage resource location and the demand destination> the storage difference> the storage age of the stored resources> the number of factories involved in the storage plan, where the number of county-level factories is constant at one; For in-transit orders, you can use the warehouse resources. The number of factories involved in the warehouse plan > the warehouse difference > the time of in-transit orders. For the agreed share of profit-making resources, the contract signing time > contract execution progress > profit difference.

5. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: The maximum matching threshold of available warehouse resources is preset in the warehouse optimization module. When the number of warehouses where available warehouse resources for a certain material match in the warehouse plan exceeds the maximum matching threshold, the warehouse plan is an inferior solution; the threshold is 3.

6. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: The economic value judgment conditions are set in the profit plan optimization module. When a profit plan is generated, the inventory cost reduction of the profit plan and the transportation cost generated by the profit transfer are compared: If the inventory cost reduced by the stock-sharing plan is greater than or equal to the transportation cost incurred by the stock-sharing transfer, then the stock-sharing plan is the optimal solution; If the inventory cost reduced by the inventory transfer plan is less than the transportation cost incurred by the inventory transfer, then the inventory transfer plan is an inferior solution.

7. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: The minimum benefit unit threshold for benefit resources is also set in the benefit plan optimization module. When the quantity of a certain material in the benefit data submodule is less than the minimum benefit unit threshold, the warehouse will not participate in the benefit.

8. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: The electric power material demand plan generation system with multi-objective resource library optimization function is compatible with the ERP system. The full material resource library is a resource library formed by resources that meet the resource library requirements in the ERP system. The resource library data submodule captures resource information within the resource library range from the ERP system through the data interface and temporarily stores it in the resource library data submodule.

9. The power material demand plan generation system with multi-objective resource optimization function according to claim 1 is characterized in that: The resource tags of the profit library in the full material resource library have a mapping relationship. When the profit library data cache submodule retrieves the profit library resource data set for profit library algorithm calculation from the full material resource library, the old materials and replaceable material resources that have a mapping relationship with the original materials in demand are retrieved and stored in the profit library data cache submodule; In the calculation of the profit algorithm submodule, different profit priority levels of profitable resources within the same profit range are considered, and profit is performed according to the following priority: complete profit of old materials > complete profit of original materials > complete profit of replaceable materials > partial profit of old materials > partial profit of original materials.

Images