Electricity meter supply chain monitoring and management method and equipment based on demand forecasting

By establishing a power meter demand forecast model and artificial intelligence algorithm, identifying supply chain breakage links and optimizing inventory and replenishment, the problems of low efficiency and inaccurate prediction of power meter supply chain in the existing technology are solved, and the efficiency of the supply chain is achieved Manage and respond quickly.

CN113554291BActive Publication Date: 2024-11-05STATE GRID SHANGHAI MUNICIPAL ELECTRIC POWER CO +1
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202110793546.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-07-14
Publication Date
2024-11-05
Estimated Expiration
2041-07-14

AI Technical Summary

Technical Problem

The existing power meter supply chain management has problems such as low manual reporting efficiency, unsatisfactory demand forecasts and weak link interconnection, resulting in duplicate equipment procurement, backlog or exhaustion of inventory, resulting in waste of costs and inability to respond high-quality and fast.

Method used

Adopt the power meter supply chain monitoring and management method based on demand forecasting, by establishing an electricity meter demand forecast model, obtaining historical installation data, performing data preprocessing and feature engineering, using artificial intelligence algorithms to predict future demand, identifying supply chain breakage links and Generate early warning signals, adjust inventory and replenishment volume, and optimize supply chain management.

Benefits of technology

It improves the efficiency of supply chain operation, reduces inventory costs, ensures rapid response of electricity meters and improves supply chain quality, reduces the cumbersome data sorting work of business personnel, optimizes the process, improves efficiency and on-time delivery rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113554291B_ABST
    Figure CN113554291B_ABST
Patent Text Reader

Abstract

The present invention relates to a method and device for monitoring and managing an electric energy meter supply chain based on demand forecasting, the method comprising the following steps: establishing an electric energy meter demand forecasting model; obtaining electric energy meter planning data, and obtaining demand forecast values ​​based on the electric energy meter demand forecasting model; judging whether there is a supply chain rupture risk based on the demand forecast values, identifying the supply chain rupture links, and generating early warning signals. Compared with the prior art, the present invention has the advantages of improving efficiency and response speed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of electric energy meter resource control, and in particular to an electric energy meter supply chain monitoring and management method and device based on demand quantity prediction. Background Art

[0002] With the increasing popularity of digital technology and the advent of digital transformation, more and more data and intelligent models can assist in making better operational decisions.

[0003] For the uncertainty and impact of external factors in business processes, risk measurement is basically collected manually and risks and trends are evaluated and analyzed.

[0004] This model usually has the following two problems:

[0005] (1) Low efficiency of manual reporting

[0006] Business personnel do not consider factors comprehensively enough and mostly rely on their own management experience and subjective judgment to submit offline demands, which is inefficient.

[0007] (2) Unsatisfactory demand forecast

[0008] In addition to being affected by policy or temporary factors, demand forecasts also have large deviations in the accuracy of forecast results.

[0009] (3) Weak interconnectivity between links

[0010] The demand forecasting link is not closely linked to the subsequent business links, which can easily lead to duplicate equipment purchases, production center pressure, inventory backlogs or depletion, and difficulty in utilizing old inventory, ultimately resulting in cost waste and the inability to provide high-quality and fast response distribution. Summary of the invention

[0011] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide an electric energy meter supply chain monitoring and management method and equipment based on demand forecasting that improves efficiency and response speed.

[0012] The purpose of the present invention can be achieved by the following technical solutions:

[0013] A method for monitoring and managing an electric energy meter supply chain based on demand forecasting includes the following steps:

[0014] Establish an energy meter demand forecasting model;

[0015] Acquire electric energy meter planning data, and acquire demand forecast value based on the electric energy meter demand forecast model;

[0016] Based on the demand forecast value, it is determined whether there is a risk of supply chain disruption, and the supply chain disruption link is identified to generate an early warning signal.

[0017] Furthermore, the electric energy meter demand prediction model is established by the following steps:

[0018] Get historical installation data;

[0019] Performing data preprocessing on the historical installation data;

[0020] Extract the features that affect the change in installation volume through feature engineering;

[0021] Artificial intelligence algorithms are used to learn and predict the future demand for electricity meter material codes, and a demand prediction model for electricity meters is constructed.

[0022] Furthermore, the data preprocessing includes data cleaning, data processing, data calculation and data storage.

[0023] Furthermore, the supply chain disruption links include reasonable security inventory disruption, agreement inventory disruption, placed order disruption, in-transit arrival disruption, inspection disruption, qualified inventory disruption and distribution disruption.

[0024] Furthermore, the method further comprises:

[0025] The inventory is adjusted based on the demand forecast value, including inventory upper limit adjustment and replenishment quantity adjustment. The formula used for inventory upper limit adjustment is:

[0026] Max=SafetyStock+D pred +Z 0.95

[0027] Among them, Max is the upper limit of inventory, SafetyStock is the safety stock value, and D pred is the demand forecast value, Z 0.95 It is the 0.95 quantile corresponding to the prediction error of the electricity meter demand prediction model.

[0028] Furthermore, the replenishment quantity adjustment is specifically as follows:

[0029] Taking the monthly arrival cycle, monthly demand forecast value, monthly demand forecast deviation rate, preset demand satisfaction rate and weight, preset inventory backlog value and weight as data input, after being processed by the optimal model, the optimal replenishment time and replenishment quantity of materials are output.

[0030] Furthermore, the early warning signals include broken links in the supply chain, number of guarantee months and number of gaps.

[0031] The present invention also provides an electronic device, comprising:

[0032] one or more processors;

[0033] Memory; and

[0034] One or more programs stored in a memory, wherein the one or more programs include instructions for executing the power meter supply chain monitoring and management method based on demand quantity prediction as described above.

[0035] The present invention also provides a computer-readable storage medium, comprising one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs include instructions for executing the electric energy meter supply chain monitoring and management method based on demand quantity forecasting as described above.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. In view of the problems of poor connection between existing key businesses such as plan submission, arrival inspection, warehousing and distribution, stagnation of front-end business leading to obstruction of subsequent processes, insufficient sharing of business collaboration information, etc., the present invention uses big data to enable supply chain demand forecasting and node planning, builds a demand forecasting model, and reasonably and accurately predicts the future demand for electricity meters to guide the procurement, arrival, inspection, and distribution of electricity meters, assist in decision-making to improve supply chain quality management, thereby reducing inventory costs and improving asset operation management efficiency.

[0038] 2. The present invention realizes effective management and monitoring of the supply chain by predicting the future demand for electric energy meters, fundamentally solving the uncertainty and other difficulties of the supply chain itself caused by "man-made rule" and "information islands".

[0039] 3. The present invention reduces the risk of breakage and ensures smooth production of electric energy meters through full-link key factors and multi-feature and multi-target characteristics. At the same time, it can quickly respond to and meet meter demands, improve supply chain operation efficiency, and effectively reduce the tedious data sorting work of business personnel, thereby achieving the purpose of optimizing processes and improving efficiency.

[0040] 4. The present invention uses demand forecasting and inventory control models to monitor assets throughout the entire process and send early warning signals to relevant terminals, such as notifying shippers of possible delays, improving order on-time delivery rates, ensuring timely delivery of materials, reducing production delays, and flexibly and perfectly executing distribution needs without stopping.

[0041] 5. The present invention uses data visualization of early warning signals to discover key issues, determine priority decisions and resolve them, improve outage prediction capabilities, and be able to take actions based on insights gained from the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the framework principle of the present invention;

[0043] Figure 2 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0044] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0045] Example 1

[0046] refer to Figure 1 and Figure 2 As shown, this embodiment provides an electric energy meter supply chain monitoring and management method based on demand forecasting, including the following steps: establishing an electric energy meter demand forecasting model; obtaining electric energy meter planning data, and obtaining a demand forecast value based on the electric energy meter demand forecasting model; judging whether there is a supply chain disruption risk based on the demand forecast value, identifying the supply chain disruption link, and generating an early warning signal.

[0047] This method aims to solve the problems of poor connection of key businesses such as existing plan submission, arrival inspection, warehousing and distribution, obstruction of subsequent processes caused by stagnation of preceding businesses, and insufficient information sharing of business collaboration. Based on the production scheduling platform (MDS system), this method builds demand forecasting and inventory control modules, fully utilizes various business data such as procurement, arrival, inspection, and distribution, uses data to support business ideas, deeply explores the laws of business connection, provides support for procurement measurement, provides fracture analysis, old warehouse establishment, and decision-making assistance for arrival, inspection, distribution, inventory and other businesses, builds an efficient and smooth metrology supply chain process, accelerates business collaboration, information sharing, and process connectivity, and realizes closed-loop supply chain management with smart decision-making and improved quality and efficiency.

[0048] like Figure 1 As shown, the above method is implemented based on a distributed multi-layer data structure, and the distributed multi-layer data structure includes a data source, a data processing layer, a service layer and an application layer.

[0049] (1) Data source: The business data source system that provides electricity meter demand forecast analysis, including the MDS system, marketing business application system, electricity consumption information collection system, and external related data.

[0050] (2) Data processing layer: It provides the quality data processing required by the electricity meter demand prediction model, including cleaning, processing, calculation, data storage, etc.

[0051] (3) Data service layer: mainly analyzes the business types and equipment code types related to the electricity meter requirements, and provides external system data interface services.

[0052] (4) Application display layer: It provides an interface for user operation and function display. Its main functions include demand forecasting, supply chain disruption analysis, inventory control, and decision support. It can predict and control the risk of supply chain disruption in advance, thereby reducing the possibility of disruption in each link and reducing the risk of supply chain disruption.

[0053] The electricity meter demand prediction model is established through the following steps: obtaining historical installation data from a data source; performing data preprocessing on the historical installation data; extracting the features affecting installation quantity changes through feature engineering; using an artificial intelligence algorithm to learn and predict the future demand for electricity meter material codes, and constructing an electricity meter demand prediction model.

[0054] Judging whether there is a risk of supply chain disruption based on the demand forecast value can be divided into three steps:

[0055] a) Determine the reasonable safety guarantee inventory gap and whether it is necessary to replenish or arrange inspection;

[0056] b) Analyze the failure situation of each link and calculate the recommended number of months of protection for each link;

[0057] c) Analyze and locate the broken links and calculate the number of gaps while meeting the distribution needs.

[0058] The main functions that can be achieved based on the above supply chain disruption risk analysis include:

[0059] a) Display the supply chain disruption situation based on the material code dimension of the energy meter, locate the broken link, and issue a message reminder of the number of gaps, so that the auxiliary center can issue a break warning and a break alarm in advance;

[0060] b) Query according to the configuration of the number of months of protection for each link, and also support selecting the number of months of protection for each link to display the break situation;

[0061] c) Display the energy meter consumption and recommended replenishment quantity according to each business breakpoint, and provide business message reminders;

[0062] d) Use visualization technology to display the center’s supply chain disruptions and recommended replenishment quantities, etc.

[0063] The supply chain disruption links include reasonable security inventory disruption, agreement inventory disruption, placed order disruption, in-transit arrival disruption, inspection disruption, qualified inventory disruption and distribution disruption.

[0064] 1. Reasonable and safe guarantee of inventory shortage

[0065] Insufficient inventory or excess inventory is caused by deviations in demand forecasting, ordering, supplier supply, production inspection, and material storage. In order to prevent this phenomenon, it is necessary to set reasonable upper and lower limits of inventory. The minimum inventory (reasonable safety guarantee inventory) is a monitoring method to prevent insufficient inventory, and the maximum inventory (inventory upper limit) is a monitoring method to prevent excessive inventory. In addition, while balancing the risk of material backlog and the risk of supply interruption, the inventory upper limit should be as low as possible.

[0066] The input of the reasonable safety guarantee inventory break algorithm includes the electricity meter demand forecast value (guaranteed month N), service level, arrival quantity, supplier, arrival cycle and distribution demand quantity. The output is the reasonable safety guarantee inventory value, number of guarantee months, margin and gap number. The business process is divided into four steps:

[0067] a) Calculate the reasonable safety guarantee stock of each material based on the reasonable safety guarantee stock model;

[0068] b) Calculate the number of months of reasonable safety inventory support;

[0069] c) On the basis of meeting the distribution needs, calculate the reasonable security guarantee inventory gap number, determine whether there is a supply chain break, whether the auxiliary business needs to replenish the stock or arrange inspection, arrival notification, etc.;

[0070] d) Monitor the consumption of reasonable safety guarantee stocks and ensure that the stocks after consumption are not lower than the reasonable safety guarantee stocks.

[0071] Calculation formula:

[0072] Different equipment has different reasonable safety guarantee stocks. Reasonable safety guarantee stocks will be adjusted dynamically with time, equipment backlog rate, and service level. The calculation method is as follows:

[0073]

[0074] Among them: SafetyStock is the safety stock value, Z is the service level, ALT is the average delivery cycle over a period of time, DSD is the standard deviation of the material demand over a period of time, AS is the average material demand over a period of time, and LTSD is the standard deviation of the delivery cycle over a period of time.

[0075] 2. Broken Agreement Inventory

[0076] The input of the agreement inventory break algorithm includes the forecast value of the electricity meter demand (guaranteed month N), the number of winning bids, the executable ratio, the number of in-transit quantities, the inventory quantity and the number of schedulable quantities. The inventory quantity includes qualified inventory, pending verification, and new purchases. The output is the number of months of agreement inventory guarantee, agreement inventory margin and agreement inventory gap. The business process is divided into four steps:

[0077] a) Combined with the demand forecast results, based on the remaining matchable agreement inventory and the execution ratio of the winning supplier, the amount of agreement inventory that needs to be matched this month is calculated;

[0078] b) Calculate the agreed inventory margin and guarantee months based on demand forecast results, the remaining quantity of agreed inventory matching, in-transit and other related factors;

[0079] c) On the basis of meeting the distribution needs, calculate the number of agreed inventory gaps, determine whether there is a shortage, and whether the auxiliary business needs to replenish the stock or arrange inspection, arrival notification, etc.;

[0080] d) Monitor the consumption of the agreed inventory and determine whether the agreed inventory level after consumption is lower than the reasonable safety guarantee inventory level.

[0081] 3. Broken order

[0082] The input of the order break algorithm includes the forecast value of the electricity meter demand (guaranteed month N), the number of winning bids, the number of orders, the number of in-transit quantities, and the number of inventory. The output is the number of guaranteed months for orders, the margin for orders, and the number of gaps for orders. The business process is divided into three steps:

[0083] a) Combined with the demand forecast results, based on the order placement data, and taking into account factors such as in-transit and qualified inventory, the margin and guaranteed months of orders placed are calculated;

[0084] b) On the basis of meeting the distribution demand, calculate the number of gaps in the orders placed, determine whether there is a gap, and whether the auxiliary business needs to replenish the goods or arrange inspection, arrival notification, etc.;

[0085] c) Monitor the consumption of orders that have been placed and ensure that the inventory after consumption is no less than a reasonable safety guarantee inventory.

[0086] The calculation logic of the order break is as follows:

[0087] Calculate the margin of placed orders

[0088] Order margin = number of orders placed + number of goods in transit + SUM (qualified goods in stock + qualified goods overdue + overdue goods to be inspected + goods to be inspected + new goods purchased) - SUM (N-month demand forecast)

[0089] Note: The forecast quantity for N demand months needs to be configured based on actual business needs.

[0090] Determine the number of months of guarantee for the order placed

[0091] First, calculate [Total inventory of primary warehouse + In-transit + Number of orders placed] = SUM{(Qualified inventory + Qualified in-stock super + To be verified + New purchase + To be verified overdue) + In-transit + Number of orders placed};

[0092] Then, IF [Total inventory in the first-level warehouse + In-transit + Ordered quantity - Forecasted demand value for the first month in the future] is less than or equal to 0, if true, judge whether [Forecasted value for the first month in the future] is equal to 0. If it is 0, the guaranteed months is 0; if it is not 0, the guaranteed months is [Total inventory in the first-level warehouse + In-transit + Ordered quantity / Forecasted demand value for the first month in the future];

[0093] Then, IF [Total inventory in the first-level warehouse + In-transit + Ordered quantity - Forecasted demand value for the first month in the future] is greater than 0, judge whether [Total inventory in the first-level warehouse + In-transit + Ordered quantity - Forecasted demand value for the first month in the future - Forecasted demand value for the second month in the future] is less than or equal to 0. If true, the guaranteed months is 1 + (Total inventory in the first-level warehouse + In-transit + Ordered quantity - Forecasted demand value for the first month in the future) / Forecasted demand value for the second month in the future; if false, judge whether [Total inventory in the first-level warehouse + In-transit + Ordered quantity - Forecasted demand value for the first month in the future - Forecasted demand value for the second month in the future - Forecasted demand value for the third month in the future] is less than or equal to 0. If true, the guaranteed months is 2 + (Total inventory in the first-level warehouse + In-transit + Ordered quantity - Forecasted demand value for the first month in the future - Forecasted demand value for the second month in the future) / Forecasted demand value for the third month in the future;

[0094] And so on.

[0095] Judge the situation of broken ordered quantity

[0096] Whether there is a break in the order supply chain = IF (Guaranteed months of orders < n, yes, no);

[0097] Remark: It can be configured that at least n months are required for the guaranteed months and the orders in the center to enter the qualified state to ensure the distribution demand.

[0098] 4. Break in in-transit arrival

[0099] The input of the in-transit arrival break algorithm includes the forecasted demand value of electric energy meters (guaranteed month N), the winning bid quantity, the arrived quantity, the in-transit quantity, and the inventory quantity. The output is the guaranteed months of in-transit arrival, the margin of in-transit arrival, and the shortage number of in-transit arrival. The business process is divided into three steps:

[0100] a) Combine the demand forecast results, on the basis of the supply plan data, and under the condition of associating qualified inventory and other factors, count the margin of in-transit arrival and the guaranteed months;

[0101] b) On the basis of meeting the distribution demand, calculate the shortage number of in-transit arrival, judge whether there is a break, and assist whether the business needs to replenish goods or arrange verification, etc.;

[0102] c) Monitor the consumption of in-transit arrivals and determine that the inventory level after consumption should not be lower than the reasonable safety stock level.

[0103] The calculation logic for in-transit arrival breaks is as follows:

[0104] Calculate the in-transit arrival margin

[0105] In-transit arrival margin = In-transit arrival quantity + SUM(Qualified in-stock + Expired qualified in-stock + Expired pending verification + Pending verification + New purchases) - SUM(N-month demand forecast value)

[0106] Note: The N-month demand forecast value needs to be configured according to the actual business.

[0107] Determine the in-transit protection months

[0108] First, calculate [Total inventory in the first-level warehouse + In-transit] = SUM{(Qualified inventory + Expired qualified in-stock in the first-level warehouse + Pending verification + New purchases + Expired pending verification) + In-transit};

[0109] Then, IF [Total inventory in the first-level warehouse + In-transit - Demand forecast value for the next month] is less than or equal to 0, if true, determine whether [Demand forecast value for the next month] is equal to 0. If it is 0, the protection months are 0; if it is not 0, the protection months are [Total inventory in the first-level warehouse + In-transit / Demand forecast value for the next month];

[0110] Then, IF [Total inventory in the first-level warehouse + In-transit - Demand forecast value for the next month] is greater than 0, determine whether [Total inventory in the first-level warehouse + In-transit - Demand forecast value for the next month - Demand forecast value for the month after next] is less than or equal to 0. If true, the protection months are 1 + (Total inventory in the first-level warehouse + In-transit - Demand forecast value for the next month) / Demand forecast value for the month after next; if false, determine whether [Total inventory in the first-level warehouse + In-transit - Demand forecast value for the next month - Demand forecast value for the month after next - Demand forecast value for the third month after next] is less than or equal to 0. If true, the protection months are 2 + (Total inventory in the first-level warehouse + In-transit - Demand forecast value for the next month - Demand forecast value for the month after next) / Demand forecast value for the third month after next;

[0111] And so on.

[0112] Determine the purchase arrival break situation

[0113] Whether there is a break in the in-transit arrival supply chain = IF(In-transit protection months < n, Yes, No)

[0114] Note: It takes at least n months to guarantee the delivery demand by ensuring the number of months and the arrival of the center into a qualified state, which can be configured.

[0115] 5. Fracture verification

[0116] The input of the calibration break algorithm includes the predicted value of the electricity meter demand (guaranteed month N), the arrival quantity, the daily calibration parameters, the monthly calibration parameters and the inventory quantity. The output is the calibration guarantee months, the calibration margin and the calibration gap number. The business process is divided into three steps:

[0117] a) Combined with the demand forecast results, based on the verification task and factors such as qualified inventory, the verification margin and guarantee months are calculated;

[0118] b) On the basis of meeting the distribution needs, calculate the number of inspection gaps, display the number of breaks, and assist in determining whether replenishment or inspection is required;

[0119] c) Monitor and verify consumption and ensure that the inventory after consumption is not lower than the reasonable safety guarantee inventory.

[0120] The calculation logic of the fracture detection is as follows:

[0121] Calculate the verification margin

[0122] Verification margin = SUM (qualified in stock + qualified overdue in stock + overdue to be inspected + to be verified + new purchase) - SUM (N-month demand forecast)

[0123] Note: The demand forecast value for N months needs to be configured based on actual business needs

[0124] Determine the number of months of certification guarantee

[0125] First, calculate [Total Inventory of First-level Warehouse] = SUM (Qualified Inventory + Qualified Super Inventory + To Be Verified + New Purchase + To Be Verified Overdue);

[0126] Then, IF [Total inventory of primary warehouse - forecast value of demand in the first month in the future] is less than or equal to 0, if it is true, determine whether [forecast value in the first month in the future] is equal to 0. If it is 0, the number of guaranteed months is 0; if it is not 0, the number of guaranteed months is [total inventory of primary warehouse / forecast value in the first month in the future];

[0127] Then, if

Total inventory in the first-level warehouse - Forecasted demand value for the next 1st month

Total inventory in the first-level warehouse - Forecasted demand value for the next 1st month - Forecasted demand value for the next 2nd month

Total inventory in the first-level warehouse - Forecasted demand value for the next 1st month - Forecasted demand value for the next 2nd month - Forecasted demand value for the next 3rd month

[0128] And so on.

[0129] Judge the verification fracture situation

[0130] Whether the verification supply chain fracture occurs = IF (Guaranteed months < n, yes, no)

[0131] Remark: It can be configured that the guaranteed months and the center's equipment to be inspected need at least n months to enter the qualified state to ensure the distribution demand.

[0132] 6. Qualified inventory fracture

[0133] The input of the qualified inventory fracture algorithm includes the forecasted demand value of electric energy meters (guaranteed month N), the inventory to be verified, the newly purchased inventory, and the actual distribution demand quantity. The output is the guaranteed months of qualified inventory, the qualified margin, and the qualified gap number. The business process is divided into three steps:

[0134] a) Combine the demand forecast results, and on the basis of qualified inventory statistics, associate factors such as reasonable safety guaranteed inventory, and count the qualified inventory margin and guaranteed months;

[0135] b) On the basis of meeting the distribution demand, calculate the qualified inventory gap number, display the fracture quantity, and assist whether the business needs to replenish goods or arrange verification, etc.;

[0136] c) Monitor the consumption situation of qualified inventory, and judge that the inventory quantity after consumption shall not be lower than the reasonable safety guaranteed inventory quantity.

[0137] The calculation logic of qualified inventory fracture is as follows:

[0138] Calculate the qualified inventory margin, that is, the abundant quantity that the guaranteed inventory needs to have to meet the business of the previous link

[0139] Qualified inventory margin = Qualified inventory + Qualified overdue inventory in stock - Forecasted demand value for the next N months.

[0140] Note: The N-month demand forecast value needs to be configured considering the actual business;

[0141] Judge the number of months of qualified inventory guarantee

[0142] First, IF [qualified inventory value - demand forecast value for the first month in the future] is less than or equal to 0, if true, judge whether [the forecast value for the first month in the future] is equal to 0. If it is 0, the number of guarantee months is 0; if it is not 0, the number of guarantee months is [qualified inventory value / demand forecast value for the first month in the future];

[0143] Then, IF [qualified inventory value - demand forecast value for the first month in the future] is greater than 0, judge whether [qualified inventory value - demand forecast value for the first month in the future - demand forecast value for the second month in the future] is less than or equal to 0. If true, the number of guarantee months is 1 + (qualified inventory value - demand forecast value for the first month in the future) / demand forecast value for the second month in the future; if false, judge whether [qualified inventory value - demand forecast value for the first month in the future - demand forecast value for the second month in the future - demand forecast value for the third month in the future] is less than or equal to 0. If true, the number of guarantee months is 2 + (qualified inventory value - demand forecast value for the first month in the future - demand forecast value for the second month in the future) / demand forecast value for the third month in the future;

[0144] And so on.

[0145] Judge the situation of qualified inventory break

[0146] Whether there is a break in the qualified inventory supply chain = IF (the number of months of qualified inventory guarantee < n, yes, no)

[0147] Note: Through the number of guarantee months, at least n months of qualified in-stock equipment in the center are required to guarantee the distribution demand, and it can be configured.

[0148] 7. Distribution break

[0149] The input of the distribution break algorithm includes the demand forecast value of electric energy meters (guarantee month N), actual distribution volume, qualified inventory, reasonable safety guarantee inventory, distribution date, replenishment quantity, and the inventory utilization of over-aged re-inspection qualified. The output is the weekly distribution task. The specific business process is as follows: Based on the current distribution management business situation, conduct planned management and analysis on the distribution process, optimize the distribution business process, warn of the impact on the distribution rhythm, locate the key problem points, remind the business personnel to conduct early control, and ensure the smoothness of the business at the front end of the distribution business, then the distribution demand can be met.

[0150] In another embodiment, the method further includes: automated management based on the demand forecast value verification, including automation of verification links, distribution control automation and inventory control automation, and automated adjustment of inventory quantity including inventory upper limit adjustment, replenishment quantity adjustment and overage inventory management.

[0151] (1) Inventory limit

[0152] Each material has its own inventory limit value. While ensuring that the inventory volume is not lower than the safety stock, the inventory limit should be as low as possible. The inventory limit is affected by the safety stock value, demand forecast value, and the accuracy of the demand forecast model. Different materials have different inventory limit values, which are dynamically adjusted according to the demand forecast error quantile.

[0153] The formula used for upper limit adjustment is:

[0154] Max=SafetyStock+D pred +Z 0.95

[0155] Among them, Max is the upper limit of inventory, SafetyStock is the safety stock value, and D pred is the demand forecast value, Z 0.95 is the 0.95 quantile corresponding to the prediction error of the demand prediction model of the electric energy meter, indicating that the error between the predicted value of the demand prediction model and the true value is less than Z 0.95 The probability is 95%.

[0156] (2) Replenishment quantity adjustment

[0157] The specific replenishment quantity adjustment is as follows: the monthly arrival cycle, monthly demand forecast value, monthly demand forecast deviation rate, preset demand satisfaction rate and weight, preset inventory backlog value and weight are used as data input, and after being processed by the optimal model, the optimal replenishment time and replenishment quantity of the materials are output.

[0158] Calculation formula: BC = -(TD-SS)

[0159] BC: Monthly replenishment value of materials

[0160] T: Current total inventory + in-transit inventory - inventory to be shipped

[0161] D: The quantity of material required from today to the end of next month

[0162] SS: Safety stock value of the material for the next month

[0163] (3) Overage Inventory Management

[0164] For equipment with a storage age of more than 2 years, priority should be given to quickly and reasonably repurchasing it to avoid inventory backlogs and improve inventory turnover efficiency.

[0165] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0166] Example 2

[0167] This embodiment provides an electronic device, including one or more processors, a memory and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the method for monitoring and managing the electric energy meter supply chain based on demand quantity forecasting as described in Example 1.

[0168] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for monitoring and managing an electric energy meter supply chain based on demand forecasting, characterized in that: The following steps are involved: Establish a demand forecasting model for electric energy meters; Acquire electric energy meter planning data, and acquire demand forecast value based on the electric energy meter demand forecast model; Based on the demand forecast value, determine whether there is a risk of supply chain disruption, identify the supply chain disruption links, and generate early warning signals. The supply chain disruption links include reasonable security inventory disruption, agreement inventory disruption, order disruption, in-transit arrival disruption, inspection disruption, qualified inventory disruption, and distribution disruption; Adjusting the inventory based on the demand forecast value, including inventory upper limit adjustment and replenishment quantity adjustment; The formula used for adjusting the inventory upper limit is: Max=SafetyStock+D pred +Z 0.95 Among them, Max is the upper limit of inventory, SafetyStock is the safety stock value, and D pred is the demand forecast value, Z 0.95 is the 0.95 quantile corresponding to the prediction error of the electric energy meter demand prediction model; The replenishment quantity adjustment is specifically as follows: Taking the monthly arrival cycle, monthly demand forecast value, monthly demand forecast deviation rate, preset demand satisfaction rate and weight, preset inventory backlog value and weight as data input, after being processed by the optimal model, the optimal replenishment time and replenishment quantity of materials are output; Judging whether there is a risk of supply chain disruption based on the demand forecast value includes: a) Determine the reasonable safety guarantee inventory gap and whether it is necessary to replenish or arrange inspection; b) Analyze the failure situation of each link and calculate the recommended number of months of protection for each link; c) Analyze and locate the broken links and calculate the number of gaps while meeting the distribution needs; The calculation method of the reasonable safety guarantee stock is as follows: Among them: SafetyStock is the safety stock value, Z is the service level, ALT is the average delivery cycle over a period of time, DSD is the standard deviation of the material demand over a period of time, AS is the average material demand over a period of time, and LTSD is the standard deviation of the delivery cycle over a period of time.

2. The method for monitoring and managing the electric energy meter supply chain based on demand forecasting according to claim 1 is characterized in that: The electric energy meter demand prediction model is established by the following steps: Get historical installation data; Performing data preprocessing on the historical installation data; Extract the features that affect the change in installation volume through feature engineering; Artificial intelligence algorithms are used to learn and predict the future demand for electricity meter material codes, and a demand prediction model for electricity meters is constructed.

3. The method for monitoring and managing the electric energy meter supply chain based on demand forecasting according to claim 2 is characterized in that: The data preprocessing includes data cleaning, data processing, data calculation and data storage.

4. The method for monitoring and managing the electric energy meter supply chain based on demand forecasting according to claim 1, characterized in that: The early warning signals include broken links in the supply chain, number of guarantee months and number of gaps.

5. An electronic device, characterized in that: include: one or more processors; Memory; and One or more programs stored in a memory, wherein the one or more programs include instructions for executing the method for monitoring and managing the supply chain of electric energy meters based on demand forecasting as described in any one of claims 1-4.

6. A computer-readable storage medium, characterized in that: It includes one or more programs executed by one or more processors of an electronic device, and the one or more programs include instructions for executing the electric energy meter supply chain monitoring and management method based on demand quantity prediction as described in any one of claims 1-4.

Citation Information

Patent Citations

  • Power industry safety stock decision analysis method based on consumption prediction

    CN105139090A

  • Electricity meter inventory demand prediction system and method

    CN108108938A

  • Smart factory management platform system

    CN108241343A

  • Resource early warning method and device based on data monitoring and computer equipment

    CN110334849A

  • Intelligent replenishment system based on supply chain demand association

    CN110516998A