CN-122022696-A - Intelligent optimization method and system for store inventory

CN122022696ACN 122022696 ACN122022696 ACN 122022696ACN-122022696-A

Abstract

The invention belongs to the technical field of retail supply chains, and particularly relates to an intelligent optimization method and system for store inventory, wherein the method comprises the steps of including a standard daily average sales volume, a standard demand fluctuation rate, a replenishment time and a customer satisfaction rate of target commodities; calculating predicted correction sales volume, acquiring a target inventory direction, calculating a dynamic customer satisfaction rate according to the boundary constraint relation between the minimum and maximum customer satisfaction rates in the customer satisfaction rates and the target inventory direction and the customer satisfaction rate, acquiring a dynamic safety inventory, acquiring a dynamic ordering point, adding the predicted correction sales volume and the dynamic ordering point of the target commodity based on the adjustment effect of the sales volume correction factors on the dynamic safety inventory and the total predicted correction sales volume in a replenishment audit period, calculating the target inventory, calculating the replenishment order quantity, generating the replenishment order quantity of each commodity and starting a purchasing and transferring flow. The method solves the problem that the prediction and decision fracture in the prior art cannot respond to future situations.

Inventors

OU JIARONG
ZHU ZHI

Assignees

广东赢商网数据服务股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. An intelligent optimization method for store inventory, comprising: building a commodity dynamic decision file for each type of commodity in store inventory, and refining intrinsic behavior attributes of the commodity in a conventional state, wherein the intrinsic behavior attributes comprise a standard daily average sales volume, a standard demand fluctuation rate, a replenishment time and a customer satisfaction rate of a target commodity; Integrating future situation events related to the target commodity, and calculating a predicted correction sales volume based on multiplication superposition rules and reference average daily sales volume; Acquiring a target inventory direction according to a preset rule base; calculating the dynamic customer satisfaction rate through the logarithmic response function regulation according to the boundary constraint relation between the minimum and maximum customer satisfaction rates in the customer satisfaction rates and the target inventory direction and the customer satisfaction rate; multiplying the dynamic customer satisfaction rate of the target commodity, the reference demand fluctuation rate and the replenishment time to obtain a dynamic safety stock, adding the predicted correction sales volume in the replenishment time to the dynamic safety stock to obtain a dynamic ordering point, and calculating the target stock based on the adjustment effect of the predicted correction sales volume on the dynamic safety stock; based on the instability of stock quantity in transportation, combining with dynamic ordering point, current stock and target stock, calculating the quantity of replenishment orders, generating the quantity of replenishment orders for each commodity and starting purchasing and allocating process.
2. The intelligent optimization method for store inventory according to claim 1, wherein the reference daily sales volume, reference demand volatility, replenishment time, customer satisfaction rate including target commodity comprises: The method comprises the steps of taking any kind of commodity in store inventory as a target commodity, obtaining a historical daily sales sequence of the target commodity in a preset time period from a store sales data system, decomposing the historical daily sales sequence by a seasonal-trend decomposition method to obtain a daily sales trend item sequence, a seasonal periodic item sequence and a residual item sequence, recording a daily sales trend item sequence corresponding to a decision day as a decision day in the future, superposing the daily sales trend item sequence and the daily sales seasonal periodic item sequence corresponding to the decision day to obtain a reference daily sales volume, recording the reference daily sales volume, calculating a standard deviation of the residual item sequence, recording the standard deviation as a reference demand fluctuation rate, extracting the replenishment time from a store commodity data management system, and extracting a conventional customer satisfaction rate target set by enterprises for the commodity, and recording the conventional customer satisfaction rate as a customer satisfaction rate.
3. The intelligent optimization method for store inventory according to claim 1, wherein the calculating the predicted revised sales volume comprises: Collecting all future events related to target commodities from data sources such as enterprise marketing calendars, supply chain management systems and the like, sorting the future events into event tuple lists and recording the event tuple lists as a future situation sequence, wherein each situation group comprises three influence factors, namely an event type, an event starting and ending date and expected influence intensity, traversing the future situation sequence, finding out the event of which all influence time ranges cover one future replenishment time or replenishment audit period and recording the event as an active event, calculating the comprehensive sales volume correction factor of the target commodities according to the influence factors of all the active events through multiplication and superposition rules and recording the comprehensive sales volume correction factor as the sales volume correction factor, multiplying the reference daily average sales volume by the sales volume correction factor to obtain predicted sales volume after situation correction, and recording the predicted sales volume as predicted correction sales volume.
4. The intelligent optimization method for store inventory according to claim 1, wherein the dynamic customer satisfaction rate satisfies the following expression: ; In the formula, Representing a dynamic customer satisfaction rate of the target commodity; Representing customer satisfaction rates of the target commodity; Representing the sales correction factor as a real number greater than 0; 、 representing a minimum customer satisfaction rate and a maximum customer satisfaction rate in the system; Representing a natural logarithmic function; Representing a maximum function; Representing a minimum function.
5. The intelligent optimization method for store inventory according to claim 1, wherein the obtaining dynamic safety inventory comprises: The dynamic customer satisfaction rate of the target commodity is obtained, the standard normal distribution quantile of the dynamic customer satisfaction rate is calculated through a statistical algorithm and recorded as the satisfaction rate quantile, and the product of the dynamic customer satisfaction rate, the satisfaction rate quantile and the replenishment time of the target commodity is used as the dynamic safety stock of the target commodity.
6. The intelligent optimization method for store inventory according to claim 1, wherein the obtaining a dynamic order point comprises: The method comprises the steps of obtaining the replenishment time of a target commodity, calculating the sum of the predicted correction sales volume of the target commodity every day in the replenishment time of the target commodity, marking the sum as the total predicted correction sales volume, and calculating the dynamic re-ordering point of the target commodity according to the sum of the total predicted correction sales volume and the dynamic safety stock, and marking the dynamic re-ordering point as the dynamic ordering point.
7. The intelligent optimization method for store inventory according to claim 1, wherein the target inventory satisfies the following expression: ; In the formula, A target inventory representing a target commodity; Representing a replenishment audit period of the target commodity; representing the predicted correction sales volume of the t-th day in the replenishment audit period of the target commodity; A sales correction factor representing the target commodity; representing a dynamic safety stock of the target commodity.
8. The intelligent optimization method for store inventory according to claim 1, wherein the replenishment order quantity satisfies the following expression: ; In the formula, Representing the replenishment order quantity of the target commodity; a target inventory representing a target commodity; representing a current inventory; A sales correction factor representing the target commodity; representing in-transit inventory; representing the maximum function.
9. The intelligent optimization method for store inventory according to claim 1, wherein the generating the replenishment order quantity for each commodity and starting the purchase allocation process comprises: And generating the quantity of the replenishment orders for each commodity in the store inventory commodity, automatically formatting the quantity of the replenishment orders, and pushing the quantity of the replenishment orders to an ERP or a supply chain management system of an enterprise corresponding to the store so as to start a subsequent purchasing or allocating process, thereby realizing end-to-end automation.
10. An intelligent optimization system for store inventory, comprising a processor and a memory storing computer program instructions that when executed by the processor implement an intelligent optimization method for store inventory according to any one of claims 1-9.

Description

Intelligent optimization method and system for store inventory Technical Field The invention relates to the technical field of retail supply chains. More particularly, the invention relates to an intelligent optimization method and system for store inventory. Background Store inventory optimization aims to balance customer satisfaction with enterprise costs by dynamically managing inventory of goods. The traditional method relies on moving average or historical average to predict, but the method assumes that the future is simple and repeated in the past, and cannot cope with the trend and seasonal change commonly existing in commodity sales, for example, annual average stock is adopted for commodities with obvious seasonality, so that the commodity is extremely easy to cause stock break in a strong season and backlog in a light season, and serious sales loss and fund waste are caused. To solve the above-described problems, a method based on time-series decomposition is disclosed in the related art, for example, a seasonal-trend decomposition method is adopted to process the historical sales data. According to the method, a complex sales volume sequence can be disassembled into a long-term trend item, a seasonal period item and a residual error item, and the random noise is removed and the standard sales volume prediction which can reflect the internal sales rules of commodities can be obtained by predicting and superposing the trend and the seasonal. However, in the process of making inventory decisions by using the reference sales volume, the prior art generally has the fundamental problems of prediction and decision-making and cracking, and specifically, after the sales volume prediction is generated, the system still depends on manually set and fixed service level targets on key decision parameters such as subsequent safety inventory, re-ordering points and the like. Such a mechanism cannot effectively incorporate known future business scenarios, such as brand promotion days, holiday peak traffic, or competing merchandise discounting activities, into the dynamic adjustment of inventory policies. This results in that even if the prediction module knows that there is great promotion in the next week, the decision module may still calculate the conservative replenishment amount because the service level objective is unchanged, eventually leading to the loss of the critical merchandise during the promotion, missing good sales, and making the inventory optimization truly intelligent. Disclosure of Invention In order to solve the technical problem that the prior art predicts and decides to split and cannot respond to future situations, the invention provides schemes in various aspects as follows. In a first aspect, the present invention provides an intelligent optimization method for store inventory, comprising: The method comprises the steps of building a commodity dynamic decision file for each type of commodity in store inventory, extracting intrinsic behavior attributes under a commodity conventional state, including a standard daily average sales volume, a standard demand fluctuation rate, a replenishment time and a customer satisfaction rate of a target commodity, integrating future situation events related to the target commodity, calculating a predicted correction sales volume based on multiplication superposition rules and the standard daily average sales volume, acquiring a target inventory direction according to a preset rule base, calculating a dynamic customer satisfaction rate according to boundary constraint relation between the minimum and maximum customer satisfaction rates in the customer satisfaction rates and the target inventory direction and the customer satisfaction rate through logarithmic response function adjustment, multiplying the dynamic customer satisfaction rate, the standard demand fluctuation rate and the replenishment time of the target commodity to obtain a dynamic safety inventory, adding the predicted correction sales volume in the replenishment time with the dynamic safety inventory to obtain a dynamic order point, calculating the target inventory based on adjustment effect of the predicted correction sales volume on the dynamic safety inventory, calculating the instability of the target inventory volume based on the transportation path, combining the current inventory and the target inventory, calculating the order quantity, generating the replenishment order quantity and starting a replenishment flow of each commodity. The method and the system do not depend on fixed historical average sales, acquire a fit actual stock benchmark by splitting sales data, avoid stock break and backlog in a strong season, integrate future events such as sales promotion and competition activity adjustment and forecast sales, solve the problem of disconnection between prediction and actual demands, flexibly adjust customer satisfaction rate, safety stock, dynamic ordering point and target sto