Logistics warehousing management method and system

By predicting inventory demand and optimizing stock space allocation, the problem of inefficient inventory management in traditional warehouse management is solved, efficient and flexible inventory management and market response capabilities are achieved, and customer satisfaction and operational efficiency are improved.

CN118628024BActive Publication Date: 2024-11-26乐麦信息技术(杭州)有限公司
View PDF 12 Cites 0 Cited by

Patent Information

Application Number
CN202411099643.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-11-26
Estimated Expiration
2044-08-12

AI Technical Summary

Technical Problem

Traditional warehouse management methods are difficult to meet the requirements of modern logistics industry for high efficiency, low cost and high reliability. There are errors in the cargo storage and out-of-stock process, and they cannot respond to market changes in a timely manner, resulting in out-of-stock or excess inventory, affecting the overall delivery time and Customer satisfaction.

Method used

By obtaining historical inventory information and market data, use seasonal ARIMA models to predict demand inventory, set reorder points and send reminders; when goods are in stock, they are allocated according to the predicted sales volume and correlation degree; when goods are out of stock, they are based on the order Picking time optimizes the order of delivery and improves picking efficiency.

Benefits of technology

Significantly reduce out-of-stock or surplus inventory, optimize inventory management, improve operational efficiency, increase sales, improve customer satisfaction, enhance warehouses' flexible response to market demand, and reduce operating costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118628024B_ABST
    Figure CN118628024B_ABST
Patent Text Reader

Abstract

The present invention proposes a logistics warehousing management method and system, wherein the method comprises: obtaining historical inventory information, predicting demand inventory according to the historical inventory information, obtaining a reorder point according to the predicted demand inventory, and sending an alarm message to the management personnel when the actual inventory reaches or falls below the reorder point to remind them to place an order; after placing an order, obtaining a warehouse entry form of a logistics warehousing warehouse of daily necessities, predicting the sales of the goods in the warehouse entry form, and allocating storage space for goods with high sales volume according to the predicted sales of the goods; obtaining information on different goods being browsed, added to shopping carts, and purchased, obtaining the goods correlation according to the information, and allocating storage space for goods with a large correlation with goods with high sales volume according to the goods correlation; when the goods are shipped out, determining the order of shipping out of the goods according to the order picking time within a preset time interval.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention proposes a logistics warehousing management method and system, belonging to the technical field of intelligent warehousing. Background Art

[0002] In recent years, with the booming development of global e-commerce and the growing market demand, traditional warehouse management methods have been unable to meet the requirements of the modern logistics industry for high efficiency, low cost and high reliability. In addition, with the advancement of technology, customers have higher expectations for the speed, accuracy and traceability of goods processing. The traditional warehouse management method has the following problems: when reordering goods, it is often determined according to previous human experience, which is prone to errors and cannot cater to market changes. When goods are put into storage, the possible future sales volume of the goods cannot be considered for storage, which will cause the goods stored in the innermost part to frequently enter and exit the warehouse, resulting in resource consumption when leaving the warehouse, or the goods cannot be stored according to the correlation between the goods. Some goods may seem unrelated, but when users buy goods, they may often buy another product because of platform promotions or group-order activities, or traditional warehouse shipments process orders according to the time when the customer places an order, resulting in slow order processing and affecting the overall average delivery time. Summary of the invention

[0003] The present invention provides a logistics warehousing management method and system to solve the above-mentioned problems:

[0004] The present invention provides a logistics warehousing management method, the method comprising:

[0005] Obtain historical inventory information, predict demand inventory based on the historical inventory information, obtain a reorder point based on the predicted demand inventory, and send an alarm message to a manager to remind him to place an order when the actual inventory reaches or falls below the reorder point;

[0006] After placing an order, obtain the incoming order of the daily necessities logistics storage warehouse, predict the sales of the goods in the incoming order, and allocate storage space for goods with high sales volume according to the predicted sales of the goods;

[0007] Obtain information about different goods being browsed, added to shopping carts, and purchased, obtain the goods relevance based on the information, and allocate storage locations for goods with high relevance to goods with high sales volume based on the goods relevance;

[0008] When the goods are shipped out, the order of shipping out is determined according to the order picking time within the preset time interval.

[0009] Furthermore, historical inventory information is obtained, demand inventory is predicted based on the historical inventory information, a reorder point is obtained based on the predicted demand inventory, and when the actual inventory reaches or falls below the reorder point, an alarm message is sent to the management personnel to remind them to place an order, including:

[0010] Acquire historical inventory information, the historical inventory information including: historical inventory data, promotion activity data of corresponding goods, and seasonal factors;

[0011] The acquired information is collated and cleaned, and future inventory demand is predicted through seasonal ARIMA;

[0012] The prediction of future inventory demand through seasonal ARIMA includes:

[0013] Divide historical data into training set and validation set;

[0014] Use the training set data to train the initial demand forecast model;

[0015] Use the initial demand forecasting model to forecast the validation set data, modify the model according to the forecast results to obtain the demand forecasting model, use the initial demand forecasting model to forecast the validation set, and obtain the predicted inventory demand quantity D j ;

[0016] The reorder point is calculated based on the predicted value. The reorder point calculation model is:

[0017]

[0018] Where R is the reorder point of the goods, μ is the predicted average demand quantity of the goods, LT is the average delivery time, Z is the safety factor of the goods corresponding to the required service level, n is the number of forecast points, and D i is the actual inventory demand quantity, D j To forecast inventory requirements;

[0019] When the inventory level of goods reaches or falls below the reorder point, an alarm message is sent to the management personnel to remind them to place an order.

[0020] Furthermore, after placing an order, obtaining a warehouse entry form of the daily necessities logistics storage warehouse, predicting the sales of the goods in the warehouse entry form, and allocating a cargo space for the goods according to the predicted sales of the goods, including:

[0021] After placing an order, obtain the incoming order, obtain the types of goods in the incoming order, and obtain the historical sales data, inventory data, and seasonal change data for each product;

[0022] Forecast sales of each product based on historical sales data, inventory data, and seasonal variation data;

[0023] Sort the goods in descending order based on the predicted sales volume;

[0024] Sort the locations in ascending order based on the picking distance from the starting picking point to the location;

[0025] The first 50% of the goods will be allocated to the sorted cargo locations in sequence.

[0026] Further, information of different goods being browsed, added to shopping carts, and purchased is obtained, and the goods relevance is obtained according to the information, and a shelf is allocated to goods with a high relevance to goods with high sales volume according to the goods relevance, including:

[0027] Obtain the information of different goods in the logistics warehouse being browsed, added to the shopping cart, and purchased, and calculate the goods relevance through the goods relevance model;

[0028] The cargo association model is:

[0029]

[0030] Among them, I(i:j t ) is the correlation between goods i and goods j at time t, x i , x j represents the specific behaviors of goods i and goods j, respectively, including purchase, browsing and adding to shopping cart, p(x i , x j , t) represents the simultaneous occurrence of cargo i in state x at time t i and good j is in state x j The probability, p(x i , t) represents that at time t, the commodity i is in state x i The independent probability of occurrence, p(x j , t) represents that at time t, the commodity j is in state x j The independent probability of occurrence;

[0031] Obtain the correlation between the goods ranked in the bottom 50% of the predicted sales volume and the goods ranked in the top 50% of the predicted sales volume;

[0032] Sort the predicted sales volume of the top 50% of goods from high to low, sort the predicted sales volume of the bottom 50% of goods based on their correlation with the top 50% of goods, obtain the bottom 50% of goods with the greatest correlation with the top 50% of goods, and assign them to the vacant storage locations closest to the top 50% of goods; when the goods with the greatest correlation have been assigned storage locations, obtain the goods with the greatest correlation that have not been assigned storage locations, and assign them to the storage locations closest to the top 50% of goods.

[0033] Furthermore, when the goods are shipped out, the order of shipping out the goods is determined according to the order picking time within a preset time interval, including:

[0034] When multiple orders are received within a preset time interval, obtain the picking time for each order;

[0035] Deliver the orders within the preset time interval in the order of least order picking time to most order picking time;

[0036] The transport instruction is sent to the picking device, and the picking device picks the goods in sequence.

[0037] The present invention proposes a logistics warehousing management system, the system comprising:

[0038] An ordering module is used to obtain historical inventory information, predict demand inventory based on the historical inventory information, obtain a reorder point based on the predicted demand inventory, and send an alarm message to the management personnel to remind them to place an order when the actual inventory reaches or falls below the reorder point;

[0039] The sales allocation cargo space module is used to obtain the incoming order of the daily necessities logistics storage warehouse after placing an order, predict the sales of the goods in the incoming order, and allocate cargo space for goods with high sales volume according to the predicted sales of the goods;

[0040] The module for associating and allocating cargo locations is used to obtain information about different goods being browsed, added to shopping carts, and purchased, obtain the degree of association of goods based on the information, and allocate cargo locations for goods with high association with goods with high sales volume based on the degree of association of goods;

[0041] The outbound module is used to determine the order of goods outbound according to the order picking time within a preset time interval when goods are outbound.

[0042] Furthermore, the ordering module includes:

[0043] A historical information acquisition module is used to acquire historical inventory information, wherein the historical inventory information includes: historical inventory data, promotion activity data of corresponding goods, and seasonal factors;

[0044] The forecasting module is used to organize and clean the acquired information and forecast future inventory demand through seasonal ARIMA. The forecasting of future inventory demand through seasonal ARIMA includes:

[0045] Divide historical data into training set and validation set;

[0046] Use the training set data to train the initial demand forecast model;

[0047] Use the initial demand forecasting model to forecast the validation set data, modify the model according to the forecast results to obtain the demand forecasting model, use the initial demand forecasting model to forecast the validation set, and obtain the predicted inventory demand quantity D j ;

[0048] The reorder point is calculated based on the predicted value. The reorder point calculation model is:

[0049]

[0050] Where R is the reorder point of the goods, μ is the predicted average demand quantity of the goods, LT is the average delivery time, Z is the safety factor of the goods corresponding to the required service level, n is the number of forecast points, and D i is the actual inventory demand quantity, D j To forecast inventory requirements;

[0051] The order reminder module is used to send alarm information to managers to remind them to place orders when the inventory level of goods reaches or falls below the reorder point.

[0052] Furthermore, the sales allocation cargo location module includes:

[0053] The module for obtaining sales data of each commodity is used to obtain the warehouse entry list after placing an order, obtain the type of goods in the warehouse entry list, and obtain the historical sales data, inventory data and seasonal change data of each commodity;

[0054] The sales forecasting module is used to forecast the sales of each product based on historical sales data, inventory data, and seasonal change data;

[0055] Sales sorting module, used to sort goods in descending order according to predicted sales volume;

[0056] The distance sorting module is used to sort the storage locations in ascending order according to the picking distance from the starting picking point to the storage location;

[0057] The allocation module is used to allocate the first 50% of the goods to the sorted cargo positions in sequence.

[0058] Furthermore, the associated allocation cargo location module includes:

[0059] The goods association calculation module is used to obtain the information of different goods being browsed, added to the shopping cart and purchased in the logistics warehouse, and calculate the goods association through the goods association model;

[0060] The cargo association model is:

[0061]

[0062] Among them, I(i; jt ) is the correlation between goods i and goods j at time t, x i , x j represents the specific behaviors of goods i and goods j, respectively, including purchase, browsing and adding to shopping cart, p(x i , x j , t) represents the simultaneous occurrence of cargo i in state x at time t i and good j is in state x j The probability, p(x i , t) represents that at time t, the commodity i is in state x i The independent probability of occurrence, p(x j , t) represents that at time t, the commodity j is in state x j The independent probability of occurrence;

[0063] A sales volume correlation module is used to obtain the correlation between the goods ranked in the bottom 50% of the predicted sales volume and the goods ranked in the top 50% of the predicted sales volume;

[0064] The sorting and allocation storage location module is used to sort the predicted sales volume of the top 50% of goods from high to low, sort the predicted sales volume of the bottom 50% of goods based on their correlation with the top 50% of goods, obtain the goods in the bottom 50% with the greatest correlation with the goods in the top 50%, and allocate them to the vacant storage location closest to the goods in the top 50%. When the goods with the greatest correlation have been allocated a storage location, obtain the goods with the greatest correlation that have not been allocated a storage location, and allocate them to the storage location closest to the goods in the top 50%.

[0065] Suppose there are two products: product i and product j. In a certain time period t, users can perform the following three operations on these two products: purchase (B), browse (V) and add to cart (A).

[0066] In this scenario, all possible state combinations of product i and product j are as follows:

[0067] (X i =B,X j =B): Both commodity i and commodity j are purchased, (X i =B,X j =V): Product i is purchased, product j is browsed, (X i =B,X j =A): Product i is purchased, product j is added to the shopping cart, (X i =V,X j =B) Product i is browsed, product j is purchased, (X i =V,X j = V): Both product i and product j are browsed, (X i=V,X j =A) Product i is browsed, product j is added to the shopping cart, (X i =A,X j =B): Product i is added to the shopping cart, and product j is purchased, (X i =A,X j =V): Product i is added to the shopping cart, product j is browsed, (X i =A,X j =A): Product i and product j are both added to the shopping cart. We need to calculate the probability P(X i ,X j ,t) and the individual probabilities P(X i ,t) and P(X j ,t), and then substitute these values ​​into the formula for calculation and sum.

[0068] Furthermore, in a logistics warehousing management system, the outbound module includes:

[0069] An order picking time acquisition module is used to acquire the picking time of each order when multiple orders are received within a preset time interval;

[0070] The picking time sorting module is used to ship the orders within the preset time interval in the order of order picking time from least to most;

[0071] The instruction sending module is used to send the transport instruction to the picking device, and the picking device picks the goods in sequence.

[0072] The beneficial effects of the present invention are as follows: inventory management optimization, accurate demand forecasting and timely replenishment can significantly reduce the situation of out-of-stock or excess inventory, thereby reducing capital occupation and warehousing costs; improving operational efficiency, by optimizing cargo space allocation and intelligent picking strategies, it can reduce the time employees spend moving around in the warehouse, speed up the overall cargo processing speed, and improve work efficiency; increasing sales, ensuring sufficient inventory and rapid delivery of hot-selling goods and related goods, can improve customer satisfaction, reduce the loss of sales opportunities, and thus may increase overall sales; flexibility in responding to market demand, the system's rapid response capability allows warehouses to flexibly adapt to changes in market demand, especially during peak promotion periods (such as supply chain disruptions); in general, the technical solution greatly improves the intelligence level of inventory management, reduces operating costs, and improves customer service quality by comprehensively utilizing data analysis, automation systems, and intelligent algorithms. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A schematic diagram of a logistics warehousing management method according to the present invention;

[0074] Figure 2 This is a schematic diagram of a logistics warehousing management system described in the present invention. DETAILED DESCRIPTION

[0075] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0076] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0077] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0078] One embodiment of the present invention, 1. A logistics warehousing management method, characterized in that the method comprises:

[0079] Acquire historical inventory information, predict demand inventory based on the historical inventory information, obtain a reorder point based on the predicted demand inventory, and send an alarm message to a manager to remind him to place an order when the actual inventory reaches or falls below the reorder point;

[0080] After placing an order, obtain the incoming order of the daily necessities logistics storage warehouse, predict the sales of the goods in the incoming order, and allocate storage space for goods with high sales volume according to the predicted sales of the goods;

[0081] Obtain information about different goods being browsed, added to shopping carts, and purchased, obtain the goods relevance based on the information, and allocate storage locations for goods with high relevance to goods with high sales volume based on the goods relevance;

[0082] When the goods are shipped out, the order of shipping out is determined according to the order picking time within the preset time interval.

[0083] The working principle and effect of the above technical solution are as follows: This technical solution combines historical inventory information, demand forecasting, reorder point setting, sales forecasting, cargo positioning and order processing strategies to form a comprehensive inventory and logistics management system, historical inventory data analysis and demand forecasting, based on past inventory data (such as past sales, seasonal fluctuations, market trends, etc.), the system uses machine learning models to predict future inventory needs, reorder point determination, based on the predicted demand inventory, the system sets the reorder point. When the actual inventory level drops to this point or lower, the system automatically issues an alarm message to remind management personnel to replenish the goods; goods warehousing and sales forecasting, when the goods arrive at the warehouse, the warehousing information is obtained, and the future sales of these goods are predicted. This prediction can be based on historical sales data, market trends, and promotional information; storage location allocation optimization, allocating high-quality storage locations to goods with high sales forecasts to facilitate rapid delivery. At the same time, the correlation between goods (such as goods that are often purchased together) is considered to determine the storage location, so as to optimize the picking path and improve picking efficiency; intelligent picking and delivery management, within the set time interval, the system proceeds according to the picking time of each order, giving priority to orders with short picking times, thereby improving overall delivery efficiency. Inventory management optimization, accurate demand forecasting and timely replenishment can significantly reduce out-of-stock or excess inventory, thereby reducing capital occupation and warehousing costs; improving operational efficiency, by optimizing storage space allocation and intelligent picking strategies, it can reduce employee movement time in the warehouse, speed up the overall cargo processing speed, and improve work efficiency; increasing sales, ensuring sufficient inventory and rapid delivery of hot-selling goods and related goods, can improve customer satisfaction, reduce the loss of sales opportunities, and thus may increase overall sales; flexibility in responding to market demand, the system's rapid response capabilities allow warehouses to flexibly adapt to changes in market demand, especially during peak promotion periods (such as supply chain disruptions); in general, this technical solution has greatly improved the intelligence level of inventory management, reduced operating costs, and improved customer service quality through the comprehensive use of data analysis, automation systems, and intelligent algorithms.

[0084] One embodiment of the present invention is a logistics warehousing management method, characterized in that historical inventory information is obtained, demand inventory is predicted based on the historical inventory information, a reorder point is obtained based on the predicted demand inventory, and when the actual inventory reaches or falls below the reorder point, an alarm message is sent to a manager to remind him to place an order, including:

[0085] Acquire historical inventory information, the historical inventory information including: historical inventory data, promotion activity data of corresponding goods, and seasonal factors;

[0086] The acquired information is collated and cleaned, and future inventory demand is predicted through seasonal ARIMA;

[0087] The prediction of future inventory demand through seasonal ARIMA includes:

[0088] Divide historical data into training set and validation set;

[0089] Use the training set data to train the initial demand forecast model;

[0090] Use the initial demand forecasting model to forecast the validation set data, modify the model according to the forecast results to obtain the demand forecasting model, use the initial demand forecasting model to forecast the validation set, and obtain the predicted inventory demand quantity D j ;

[0091] The reorder point is calculated based on the predicted value. The reorder point calculation model is:

[0092]

[0093] Where R is the reorder point of the goods, μ is the predicted average demand quantity of the goods, LT is the average delivery time, Z is the safety factor of the goods corresponding to the required service level, n is the number of forecast points, and D i is the actual inventory demand quantity, D j To forecast inventory requirements;

[0094] When the inventory level of goods reaches or falls below the reorder point, an alarm message is sent to the management personnel to remind them to place an order.

[0095] The working principle and effect of the above technical solution are as follows: this technical solution uses the seasonal ARIMA (AutoRegressive Integrated Moving Average) model to implement inventory management and optimization, data collection and preprocessing, first collect historical inventory data, and combine relevant promotional activity data and seasonal factors to organize and clean the collected data, eliminate noise and outliers, and ensure data quality; model construction and training, divide the cleaned data into a training set and a validation set, and use the training set data to train the seasonal ARIMA model. This model can capture seasonal fluctuations and trends in the data and is suitable for predicting time series data with obvious periodic patterns; demand forecasting and verification, use the trained model to predict the validation set to evaluate the accuracy and reliability of the model and predict the inventory demand value at each time point in the future; calculation of the reorder point is based on the demand predicted by the model and the parameters estimated from historical data (such as average demand μ, average delivery time, etc.), calculate the reorder point, inventory monitoring and management, the system monitors the real-time inventory level, once the inventory is lower than the calculated reorder point, immediately send an alarm message to the management personnel to prompt replenishment operations. Accuracy is improved. The seasonal ARIMA model can be used to more accurately predict inventory demand with cyclical and trend characteristics, making inventory management more scientific and precise. Timely response to market changes. By monitoring real-time inventory and comparing it with the preset reorder point, the system can quickly respond to market changes and reduce losses caused by out-of-stock or excess inventory. Cost optimization. Forecasting demand and optimizing inventory levels in advance can significantly reduce warehousing costs and prevent funds from being idle in excessive inventory. Service levels are improved. Maintaining appropriate inventory levels ensures that customer needs can be met and avoids sales opportunities caused by out-of-stock, thereby improving customer satisfaction and brand reputation. Risk management. By setting the safety factor Z to adjust the reorder point, the company reflects the trade-off between out-of-stock risk and inventory cost. A high safety factor means a higher inventory level to avoid out-of-stock risk, while a lower safety factor may reduce inventory carrying costs but increase the risk of out-of-stock. Choosing an appropriate Z value can be flexibly adjusted according to market demand and corporate strategy.

[0096] One embodiment of the present invention is a logistics warehousing management method, which obtains a warehouse entry list of a logistics warehousing warehouse for daily necessities after placing an order, predicts the sales of the goods in the warehouse entry list, and allocates a cargo space for the goods according to the predicted sales of the goods, including:

[0097] After placing an order, obtain the incoming order, obtain the types of goods in the incoming order, and obtain the historical sales data, inventory data, and seasonal change data for each product;

[0098] Forecast sales of each product based on historical sales data, inventory data, and seasonal variation data;

[0099] Sort the goods in descending order based on the predicted sales volume;

[0100] Sort the locations in ascending order based on the picking distance from the starting picking point to the location;

[0101] Allocate the first 50% of the goods to the sorted cargo locations in sequence.

[0102] The working principle and effect of the above technical solution are as follows: the technical solution integrates the historical sales data, inventory data and seasonal change factors of the incoming goods and performs the following main steps: data collation and analysis, obtaining the incoming order after ordering, and listing the types of goods in detail; collecting historical sales data, inventory data and seasonal change data for each type of goods;

[0103] Sales forecasting, using historical sales data, inventory status and seasonal data, and applying statistical or machine learning models (such as time series forecasting models) to predict the future sales performance of each product; cargo sorting, sorting the types of goods in descending order according to the predicted sales volume to determine which products are expected to have higher sales; storage location sorting and allocation, sorting the storage locations in the warehouse in ascending order according to the picking distance from the starting picking point to the storage location to optimize the picking route and reduce the distance employees move; allocating the top 50% of the predicted sales volume to the top 50% of the storage locations, aiming to place high-selling products in a more convenient picking location.

[0104] Improve picking efficiency by allocating high-selling goods to more accessible locations, reduce picking time and labor costs, and improve work efficiency; optimize inventory management, accurate sales forecasts can help better manage inventory, avoid excess or out-of-stock inventory, reduce holding costs, and improve capital turnover; improve customer satisfaction, speed up the picking and delivery process, respond to customer orders more quickly, and improve customer satisfaction and service levels; reduce logistics costs, optimize the layout of cargo spaces, and reduce transportation and handling distances within the warehouse, thereby reducing internal logistics costs; flexibly respond to market changes, consider seasonal changes in cargo sales forecasts so that warehouses can adjust inventory strategies according to seasonal fluctuations in market demand and better respond to market changes such as holiday sales peaks. In general, by efficiently combining sales forecasts with physical layout optimization, this technical solution can significantly improve warehouse operating efficiency and response speed, making the entire supply chain more efficient and competitive.

[0105] One embodiment of the present invention is a logistics warehousing management method, which obtains information about different goods being browsed, added to shopping carts, and purchased, obtains the goods relevance based on the information, and allocates cargo locations for goods with a high relevance to goods with a high sales volume based on the goods relevance, including:

[0106] Obtain the information of different goods being browsed, added to the shopping cart, and purchased in the logistics warehouse, and calculate the goods relevance through the goods relevance model, the goods relevance model is:

[0107]

[0108] Among them, I(i; j t ) is the correlation between goods i and goods j at time t, x i , x j represents the specific behaviors of goods i and goods j, respectively, including purchase, browsing and adding to shopping cart, p(x i , x j , t) represents the simultaneous occurrence of cargo i in state x at time t i and good j is in state x j The probability, p(x i , t) represents that at time t, the commodity i is in state x i The independent probability of occurrence, p(x j , t) represents that at time t, the commodity j is in state x j The independent probability of occurrence;

[0109] Obtain the correlation between the goods ranked in the bottom 50% of the predicted sales volume and the goods ranked in the top 50% of the predicted sales volume;

[0110] Sort the predicted sales volume of the top 50% of goods from high to low, sort the predicted sales volume of the bottom 50% of goods based on their correlation with the top 50% of goods, obtain the bottom 50% of goods with the greatest correlation with the top 50% of goods, and assign them to the vacant storage locations closest to the top 50% of goods; when the goods with the greatest correlation have been assigned storage locations, obtain the goods with the greatest correlation that have not been assigned storage locations, and assign them to the storage locations closest to the top 50% of goods.

[0111] The goods association model includes various combinations of states or behaviors. Goods i and j can be in a variety of different states at a given time t (such as purchase, browse, add to cart, etc.). The sum is to ensure that the algorithm can include all possible state combinations of these goods and calculate the association between them; Comprehensive evaluation of overall mutual information: Mutual information is a measure of the shared information between two variables. Summing all possible state combinations means that the amount of information of goods i and j under all possible interactions is comprehensively evaluated, which provides a comprehensive perspective to understand how different states or behaviors affect each other as a whole within time t; Calculate the overall association. Evaluating each state or behavior separately may not be enough to express the comprehensive relationship between goods. For example, considering the "purchase" behavior alone may ignore the interaction of "browse" or "add to cart" behaviors. Comprehensive all behavior patterns provide a more comprehensive correlation analysis.

[0112] Suppose there are two products: product i and product j. In a certain time period t, users can perform the following three operations on these two products: purchase (B), browse (V) and add to cart (A).

[0113] In this scenario, all possible state combinations of product i and product j are as follows:

[0114] (X i =B,X j =B): Both commodity i and commodity j are purchased, (X i =B,X j =V): Product i is purchased, product j is browsed, (X i =B,X j =A): Product i is purchased, product j is added to the shopping cart, (X i =V,X j =B) Product i is browsed, product j is purchased, (X i =V,X j = V): Both product i and product j are browsed, (X i =V,X j =A) Product i is browsed, product j is added to the shopping cart, (X i =A,X j =B): Product i is added to the shopping cart, and product j is purchased, (X i =A,X j =V): Product i is added to the shopping cart, product j is browsed, (X i =A,X j =A): Product i and product j are both added to the shopping cart. We need to calculate the probability P(X i ,X j ,t) and the individual probabilities P(Xi ,t) and P(X j ,t), and then substitute these values ​​into the formula for calculation and sum.

[0115] The working principle and effect of the above technical solution are as follows: the technical solution uses a goods correlation model to optimize the layout of goods in the warehouse, data collection and analysis, collects behavioral data of different goods in the warehouse, including browsing, adding to shopping carts and purchasing; calculates the goods correlation, applies the goods correlation model, which measures the correlation between two goods by comparing the ratio between the behavioral probability of the goods and the probability of their respective independent behaviors, sorts the goods with predicted sales volume, and screens out the top 50% with higher sales volume and the bottom 50% with lower sales volume; calculates the correlation between the top 50% high-selling goods and the bottom 50% of the goods; according to the correlation and sales volume, matches the bottom 50% of the goods with the goods with the highest sales volume in the top 50% according to the correlation, and allocates the goods to the storage space; the low High-selling goods are allocated to free storage spaces near high-selling goods, thereby improving delivery efficiency and the convenience for customers to purchase a variety of goods; improving picking efficiency by placing highly related goods close to each other to reduce the moving distance during picking, thereby improving picking efficiency; optimizing warehouse space usage: reasonable goods layout makes hot-selling goods more accessible, while ensuring the logical placement of related goods and effectively utilizing warehouse space; based on dynamic adjustment of data, this solution allows regular updating and adjustment of the relevance and location of goods, giving the strategy the flexibility to dynamically adjust to cope with changes in the market and consumer behavior; in general, through intelligent analysis and strategic storage space arrangement, this technical solution not only improves the efficiency of warehouse operations, but also enhances customers' purchasing experience and increases sales. It is a warehouse management solution with comprehensive performance improvement.

[0116] One embodiment of the present invention is a logistics warehousing management method, when goods are shipped out, determining the order of goods shipped out according to order picking time within a preset time interval, including:

[0117] When multiple orders are received within a preset time interval, obtain the picking time for each order;

[0118] Deliver the orders within the preset time interval in the order of least order picking time to most order picking time;

[0119] The transport instruction is sent to the picking device, and the picking device picks the goods in sequence.

[0120] The order of goods outbound is determined according to the order picking time within the preset time interval. When there is no idle picking equipment in the warehouse, it is still based on first come first served. When there is no idle picking equipment, the time interval can be preset by referring to the following time interval model.

[0121] T=min(T b +α·ln(1+N)+β·ln(1+M), Tmax)

[0122] Where, T is the preset time interval, T b is the basic time interval, which is the minimum time interval that the system should maintain regardless of the number or size of orders. The basic time interval can be set to the average time it takes for the AGV to process an order. α is the adjustment factor related to the number of orders. N is the total number of orders currently to be processed. β is the adjustment factor related to the average number of goods in the order. M is the average number of goods in all currently pending orders. T max is the preset maximum time interval.

[0123] The working principle and effect of the above technical solution are as follows: the technical solution optimizes the order processing process, enables the warehouse to process and ship orders more efficiently, collects order information, and within a preset time interval, the system collects information on each order received, calculates and sorts the picking time, and calculates the complete picking time required for each order, which is based on factors such as the location, quantity and related picking paths of the goods in the order; sorts the collected orders from least to most according to the time required for picking; optimizes order delivery, and prioritizes orders with short picking times according to the sorting of picking times. This means that orders that require less time to prepare can be sent out first, thereby improving overall delivery efficiency; the system will automatically send the picking order to the picking equipment, such as an automated picking robot or a traditional picker; the picking equipment or personnel will perform picking operations in accordance with the instructions of the system. Improve processing speed. By optimizing the processing sequence of orders, that is, giving priority to orders with shorter picking times, the overall order processing speed is accelerated; improve customer satisfaction. Faster order processing speed usually means that customers can receive goods faster, which directly improves customer satisfaction; improve efficiency. A clear picking sequence reduces the aimless movement of equipment and personnel in the warehouse, reduces time waste, and improves the overall efficiency of picking. Reduce error rate. The system's automated instructions reduce errors caused by human factors, such as picking the wrong goods or missing certain orders; optimize resource utilization. Reasonable time sequencing and machine scheduling enable more efficient use of warehouse resources, including the use of employee time and mechanical equipment. Improve structured operations. Unified and automated processing procedures improve the level of structuring of operations and reduce confusion and uncertainty in operations. In short, this technical solution makes warehouse operations more efficient and systematic by intelligently optimizing the processing sequence of orders, significantly improves processing speed and customer service quality, and achieves higher operating efficiency and lower error rates; use the provided formula to calculate the current time interval, where T bIt is the minimum basic time interval that the system should maintain when the order volume is the smallest. The parameter α controls the impact of the number of orders (N) on the interval, while β controls the impact of the average number of goods (M). Orders within this time interval are sorted from small to large according to the cargo location information items to generate a shipping sequence list. The purpose of generating a shipping sequence list is to minimize the average shipping time when the AGV car resources are limited and cannot respond immediately to the user's order request. By sorting the orders within this time interval from small to large according to the cargo location information items, the average shipping time of the orders is significantly reduced. Because the orders with less cargo information are processed first, they will leave the system faster, resulting in a reduction in the number of orders waiting for shipment and the average shipping time of all orders. Shorter jobs can process more jobs per unit time because they are completed quickly, which will increase the system throughput. Reduce the cumulative delay of orders. Early completion of orders means that subsequent tasks that depend on these orders can start earlier, shortening the cumulative delay in the order process. Improve user experience. For users, seeing a fast-responding system will give them a better experience, which is especially important for applications that require interactive responses.

[0124] One embodiment of the present invention is a logistics warehousing management system, the system comprising:

[0125] An ordering module is used to obtain historical inventory information, predict demand inventory based on the historical inventory information, obtain a reorder point based on the predicted demand inventory, and send an alarm message to the management personnel to remind them to place an order when the actual inventory reaches or falls below the reorder point;

[0126] The sales allocation cargo space module is used to obtain the incoming order of the daily necessities logistics storage warehouse after placing an order, predict the sales of the goods in the incoming order, and allocate cargo space for goods with high sales volume according to the predicted sales of the goods;

[0127] The module for associating and allocating cargo locations is used to obtain information about different goods being browsed, added to shopping carts, and purchased, obtain the degree of association of goods based on the information, and allocate cargo locations for goods with high association with goods with high sales volume based on the degree of association of goods;

[0128] The outbound module is used to determine the order of goods outbound according to the order picking time within a preset time interval when goods are outbound.

[0129] The working principle and effect of the above technical solution are as follows: This technical solution combines historical inventory information, demand forecasting, reorder point setting, sales forecasting, cargo positioning and order processing strategies to form a comprehensive inventory and logistics management system, historical inventory data analysis and demand forecasting, based on past inventory data (such as past sales, seasonal fluctuations, market trends, etc.), the system uses machine learning models to predict future inventory needs, reorder point determination, based on the predicted demand inventory, the system sets the reorder point. When the actual inventory level drops to this point or lower, the system automatically issues an alarm message to remind management personnel to replenish the goods; goods warehousing and sales forecasting, when the goods arrive at the warehouse, the warehousing information is obtained, and the future sales of these goods are predicted. This prediction can be based on historical sales data, market trends, and promotional information; storage location allocation optimization, allocating high-quality storage locations to goods with high sales forecasts to facilitate rapid delivery. At the same time, the correlation between goods (such as goods that are often purchased together) is considered to determine the storage location, so as to optimize the picking path and improve picking efficiency; intelligent picking and delivery management, within the set time interval, the system proceeds according to the picking time of each order, giving priority to orders with short picking times, thereby improving overall delivery efficiency. Inventory management optimization, accurate demand forecasting and timely replenishment can significantly reduce out-of-stock or excess inventory, thereby reducing capital occupation and warehousing costs; improving operational efficiency, by optimizing storage space allocation and intelligent picking strategies, it can reduce employee movement time in the warehouse, speed up the overall cargo processing speed, and improve work efficiency; increasing sales, ensuring sufficient inventory and rapid delivery of hot-selling goods and related goods, can improve customer satisfaction, reduce the loss of sales opportunities, and thus may increase overall sales; flexibility in responding to market demand, the system's rapid response capabilities allow warehouses to flexibly adapt to changes in market demand, especially during peak promotion periods (such as supply chain disruptions); in general, this technical solution has greatly improved the intelligence level of inventory management, reduced operating costs, and improved customer service quality through the comprehensive use of data analysis, automation systems, and intelligent algorithms.

[0130] In one embodiment of the present invention, a logistics warehousing management system, the ordering module includes:

[0131] A historical information acquisition module is used to acquire historical inventory information, wherein the historical inventory information includes: historical inventory data, promotion activity data of corresponding goods, and seasonal factors;

[0132] The forecasting module is used to organize and clean the acquired information and forecast future inventory demand through seasonal ARIMA. The forecasting of future inventory demand through seasonal ARIMA includes:

[0133] Divide historical data into training set and validation set;

[0134] Use the training set data to train the initial demand forecast model;

[0135] Use the initial demand forecasting model to forecast the validation set data, modify the model according to the forecast results to obtain the demand forecasting model, use the initial demand forecasting model to forecast the validation set, and obtain the predicted inventory demand quantity D j ;

[0136] The reorder point is calculated based on the predicted value. The reorder point calculation model is:

[0137]

[0138] Where R is the reorder point of the goods, μ is the predicted average demand quantity of the goods, LT is the average delivery time, Z is the safety factor of the goods corresponding to the required service level, n is the number of forecast points, and D i is the actual inventory demand quantity, D j To forecast inventory requirements;

[0139] The order reminder module is used to send alarm information to managers to remind them to place orders when the inventory level of goods reaches or falls below the reorder point.

[0140] The working principle and effect of the above technical solution are as follows: this technical solution uses the seasonal ARIMA (AutoRegressive Integrated Moving Average) model to implement inventory management and optimization, data collection and preprocessing, first collect historical inventory data, and combine relevant promotional activity data and seasonal factors to organize and clean the collected data, eliminate noise and outliers, and ensure data quality; model construction and training, divide the cleaned data into a training set and a validation set, and use the training set data to train the seasonal ARIMA model. This model can capture seasonal fluctuations and trends in the data and is suitable for predicting time series data with obvious periodic patterns; demand forecasting and verification, use the trained model to predict the validation set to evaluate the accuracy and reliability of the model and predict the inventory demand value at each time point in the future; calculation of the reorder point is based on the demand predicted by the model and the parameters estimated from historical data (such as average demand μ, average delivery time, etc.), calculate the reorder point, inventory monitoring and management, the system monitors the real-time inventory level, once the inventory is lower than the calculated reorder point, immediately send an alarm message to the management personnel to prompt replenishment operations. Accuracy is improved. The seasonal ARIMA model can be used to more accurately predict inventory demand with cyclical and trend characteristics, making inventory management more scientific and precise. Timely response to market changes. By monitoring real-time inventory and comparing it with the preset reorder point, the system can quickly respond to market changes and reduce losses caused by out-of-stock or excess inventory. Cost optimization. Forecasting demand and optimizing inventory levels in advance can significantly reduce warehousing costs and prevent funds from being idle in excessive inventory. Service levels are improved. Maintaining appropriate inventory levels ensures that customer needs can be met and avoids sales opportunities caused by out-of-stock, thereby improving customer satisfaction and brand reputation. Risk management. By setting the safety factor Z to adjust the reorder point, the company reflects the trade-off between out-of-stock risk and inventory cost. A high safety factor means a higher inventory level to avoid out-of-stock risk, while a lower safety factor may reduce inventory carrying costs but increase the risk of out-of-stock. Choosing an appropriate Z value can be flexibly adjusted according to market demand and corporate strategy.

[0141] In one embodiment of the present invention, a logistics warehousing management system, the sales allocation cargo location module includes:

[0142] The module for obtaining sales data of each commodity is used to obtain the warehouse entry list after placing an order, obtain the type of goods in the warehouse entry list, and obtain the historical sales data, inventory data and seasonal change data of each commodity;

[0143] The sales forecasting module is used to forecast the sales of each product based on historical sales data, inventory data, and seasonal variation data;

[0144] Sales sorting module, used to sort goods in descending order according to predicted sales volume;

[0145] The distance sorting module is used to sort the storage locations in ascending order according to the picking distance from the starting picking point to the storage location;

[0146] The allocation module is used to allocate the first 50% of the goods to the sorted cargo positions in sequence.

[0147] In one embodiment of the present invention, a logistics warehousing management system, the associated allocation cargo location module includes:

[0148] The goods association calculation module is used to obtain the information of different goods being browsed, added to the shopping cart and purchased in the logistics warehouse, and calculate the goods association through the goods association model. The goods association model is:

[0149]

[0150] Among them, I(i; j t ) is the correlation between goods i and goods j at time t, x i , x j represents the specific behaviors of goods i and goods j, respectively, including purchase, browsing and adding to shopping cart, p(x i , x j , t) represents the simultaneous occurrence of cargo i in state x at time t i and good j is in state x j The probability, p(x i , t) represents that at time t, the commodity i is in state x i The independent probability of occurrence, p(x j , t) represents that at time t, the commodity j is in state x j The independent probability of occurrence;

[0151] A sales volume correlation module is used to obtain the correlation between the goods ranked in the bottom 50% of the predicted sales volume and the goods ranked in the top 50% of the predicted sales volume;

[0152] The sorting and allocation storage location module is used to sort the predicted sales volume of the top 50% of goods from high to low, sort the predicted sales volume of the bottom 50% of goods based on their correlation with the top 50% of goods, obtain the goods in the bottom 50% with the greatest correlation with the goods in the top 50%, and allocate them to the vacant storage location closest to the goods in the top 50%. When the goods with the greatest correlation have been allocated a storage location, obtain the goods with the greatest correlation that have not been allocated a storage location, and allocate them to the storage location closest to the goods in the top 50%.

[0153] The working principle and effect of the above technical solution are as follows: the technical solution uses a goods correlation model to optimize the layout of goods in the warehouse, data collection and analysis, collects behavioral data of different goods in the warehouse, including browsing, adding to shopping carts and purchasing; calculates the goods correlation, applies the goods correlation model, which measures the correlation between two goods by comparing the ratio between the behavioral probability of the goods and the probability of their respective independent behaviors, sorts the goods with predicted sales volume first, and screens out the top 50% with higher sales volume and the bottom 50% with lower sales volume; calculates the correlation between the top 50% high-selling goods and the bottom 50% of the goods; according to the correlation and sales volume, matches the bottom 50% of the goods with the goods with the highest sales volume in the top 50% according to the correlation, and allocates the goods to the shelves; the low sales volume with high correlation is selected. High-selling goods are allocated to free storage spaces near high-selling goods, thereby improving delivery efficiency and the convenience for customers to purchase a variety of goods; improving picking efficiency by placing highly related goods close to each other to reduce the moving distance during picking, thereby improving picking efficiency; optimizing warehouse space usage: reasonable goods layout makes hot-selling goods more accessible, while ensuring the logical placement of related goods and effectively utilizing warehouse space; based on dynamic adjustment of data, this solution allows regular updating and adjustment of the relevance and location of goods, giving the strategy the flexibility to dynamically adjust to cope with changes in the market and consumer behavior; in general, through intelligent analysis and strategic storage space arrangement, this technical solution not only improves the efficiency of warehouse operations, but also enhances customers' purchasing experience and increases sales. It is a warehouse management solution with comprehensive performance improvement.

[0154] In one embodiment of the present invention, a logistics warehousing management system, the outbound module includes:

[0155] An order picking time acquisition module is used to acquire the picking time of each order when multiple orders are received within a preset time interval;

[0156] The picking time sorting module is used to ship the orders within the preset time interval in the order of order picking time from least to most;

[0157] The instruction sending module is used to send the transport instruction to the picking device, and the picking device picks the goods in sequence.

[0158] The order of goods outbound is determined according to the order picking time within the preset time interval. When there is no idle picking equipment in the warehouse, it is still based on first come first served. When there is no idle picking equipment, the time interval can be preset by referring to the following time interval model.

[0159] T=min(T b +α·ln(1+N)+β·ln(1+M), Tmax)

[0160] Where, T is the preset time interval, T b is the basic time interval, which is the minimum time interval that the system should maintain regardless of the number or size of orders. The basic time interval can be set to the average time it takes for the AGV to process an order. α is the adjustment factor related to the number of orders. N is the total number of orders currently to be processed. β is the adjustment factor related to the average number of goods in the order. M is the average number of goods in all currently pending orders. T max is the preset maximum time interval.

[0161] The working principle and effect of the above technical solution are as follows: the technical solution optimizes the order processing process, enables the warehouse to process and ship orders more efficiently, collects order information, and within a preset time interval, the system collects information on each order received, calculates and sorts the picking time, and calculates the complete picking time required for each order, which is based on factors such as the location, quantity and related picking paths of the goods in the order; sorts the collected orders from least to most according to the time required for picking; optimizes order delivery, and prioritizes orders with short picking times according to the sorting of picking times. This means that orders that require less time to prepare can be sent out first, thereby improving overall delivery efficiency; the system will automatically send the picking order to the picking equipment, such as an automated picking robot or a traditional picker; the picking equipment or personnel will perform picking operations in accordance with the instructions of the system. Improve processing speed. By optimizing the processing sequence of orders, that is, giving priority to orders with shorter picking times, the overall order processing speed is accelerated; improve customer satisfaction. Faster order processing speed usually means that customers can receive goods faster, which directly improves customer satisfaction; improve efficiency. A clear picking sequence reduces the aimless movement of equipment and personnel in the warehouse, reduces time waste, and improves the overall efficiency of picking. Reduce error rate. The system's automated instructions reduce errors caused by human factors, such as picking the wrong goods or missing certain orders; optimize resource utilization. Reasonable time sequencing and machine scheduling enable more efficient use of warehouse resources, including the use of employee time and mechanical equipment. Improve structured operations. Unified and automated processing procedures improve the level of structuring of operations and reduce confusion and uncertainty in operations.In summary, this technical solution makes warehouse operations more efficient and systematic by intelligently optimizing the processing sequence of orders, significantly improving processing speed and customer service quality, and achieving higher operating efficiency and lower error rate; the provided formula is used to calculate the current time interval, where Tb is the minimum basic time interval that the system should maintain when the order volume is the smallest, the parameter α controls the impact of the number of orders (N) on the interval, and β controls the impact of the average number of goods (M); the orders within this time interval are sorted from small to large according to the cargo location information items to generate a shipping sequence list. The purpose of generating a shipping sequence list is to minimize the average when the AGV car resources are limited and cannot respond immediately to the user's order request. Shipping time: By sorting the orders within this time interval according to the cargo location information items from few to many, the average shipping time of the orders can be significantly reduced. Since orders with less cargo information are processed first, they will leave the system faster, resulting in a decrease in the number of orders waiting to be shipped and the average shipping time of all orders. Shorter jobs can be completed quickly, so more jobs can be processed per unit time, which will increase system throughput. The cumulative delay of orders is reduced. Early completion of orders means that subsequent tasks that depend on these orders can start earlier, shortening the cumulative delay in the order process. Improve user experience: users will have a better experience when they see a fast-responding system, which is especially important for applications that require interactive responses.

[0162] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A logistics warehousing management method, characterized in that: The method comprises: Acquire historical inventory information, predict demand inventory based on the historical inventory information, obtain a reorder point based on the predicted demand inventory, and send an alarm message to a manager to remind him to place an order when the actual inventory reaches or falls below the reorder point; The acquiring of historical inventory information, predicting demand inventory based on the historical inventory information, acquiring a reorder point based on the predicted demand inventory, and sending an alarm message to a manager to remind him to place an order when the actual inventory reaches or falls below the reorder point, includes: Acquire historical inventory information, the historical inventory information including: historical inventory data, promotion activity data of corresponding goods, and seasonal factors; The acquired information is collated and cleaned, and future inventory demand is predicted through seasonal ARIMA; The prediction of future inventory demand through seasonal ARIMA includes: Divide historical data into training set and validation set; Use the training set data to train the initial demand forecast model; Use the initial demand forecasting model to forecast the validation set data, and modify the initial demand forecasting model according to the forecasting result to obtain a demand forecasting model; Input the current inventory information into the demand forecasting model to obtain the predicted inventory demand quantity D j ; The reorder point is calculated based on the predicted inventory demand quantity, and the reorder point calculation model is: Where R is the reorder point of the goods, μ is the predicted average demand quantity of the goods, LT is the average delivery time, Z is the safety factor of the goods corresponding to the required service level, n is the number of forecast points, and D i is the actual inventory demand quantity, D j To forecast inventory requirements; When the inventory level reaches or falls below the reorder point, an alarm message is sent to the manager to remind him to place an order; After placing an order, obtain the incoming order of the daily necessities logistics storage warehouse, predict the sales of the goods in the incoming order, and allocate storage space for goods with high sales volume according to the predicted sales of the goods; Obtain information about different goods being browsed, added to shopping carts, and purchased, obtain the goods relevance based on the information, and allocate storage locations for goods with high relevance to goods with high sales volume based on the goods relevance; The obtaining of information on browsing, adding to shopping carts and purchasing of different goods, obtaining the goods relevance according to the information, and allocating cargo spaces to goods with high relevance to goods with high sales volume according to the goods relevance, includes: Obtain information about different goods in the logistics warehouse being browsed, added to the shopping cart, and purchased, and calculate the goods relevance through the goods relevance model. The cargo association model is: Among them, I(i; j t ) is the correlation between goods i and goods j at time t, x i ,x j represents the specific behaviors of goods i and goods j, including purchase, browsing and adding to shopping cart, p(x i ,x j ,t) represents the simultaneous occurrence of cargo i in state x at time t i and good j is in state x j The probability, p(x i ,t) represents that at time t, the item i is in state x i The independent probability of occurrence, p(x j ,t) represents that at time t, the commodity j is in state x j The independent probability of occurrence; Obtain the correlation between the goods ranked in the bottom 50% of the predicted sales volume and the goods ranked in the top 50% of the predicted sales volume; Sort the predicted sales volume of the top 50% of goods from high to low, sort the predicted sales volume of the bottom 50% of goods by their relevance to the top 50% of goods, obtain the goods in the bottom 50% with the greatest relevance to the goods in the top 50%, and assign them to the vacant cargo space closest to the goods in the top 50%. When the goods with the greatest relevance have been assigned cargo spaces, obtain the goods with the greatest relevance that have not been assigned cargo spaces, and assign them to the cargo spaces closest to the goods in the top 50%. When the goods are shipped out, the order of shipping out is determined according to the order picking time within the preset time interval.

2. A logistics warehousing management method according to claim 1, characterized in that: After placing an order, obtain the incoming order of the daily necessities logistics storage warehouse, predict the sales of the goods in the incoming order, and allocate storage space for the goods according to the predicted sales of the goods, including: After placing an order, obtain the incoming order, obtain the types of goods in the incoming order, and obtain the historical sales data, inventory data, and seasonal change data for each product; Forecast sales of each product based on historical sales data, inventory data, and seasonal variation data; Sort the goods in descending order based on the predicted sales volume; Sort the locations in ascending order based on the picking distance from the starting picking point to the location; Allocate the first 50% of the goods to the sorted cargo locations in sequence.

3. A logistics warehousing management method according to claim 1, characterized in that: When the goods are shipped out, the order of shipping out is determined according to the order picking time within the preset time interval, including: When multiple orders are received within a preset time interval, obtain the picking time for each order; Deliver the orders within the preset time interval in the order of least order picking time to most order picking time; The transport instruction is sent to the picking device, and the picking device picks the goods in sequence.

4. A logistics warehousing management system, characterized in that: The system comprises: An ordering module is used to obtain historical inventory information, predict demand inventory based on the historical inventory information, obtain a reorder point based on the predicted demand inventory, and send an alarm message to the management personnel to remind them to place an order when the actual inventory reaches or falls below the reorder point; Wherein, the ordering module includes: A historical information acquisition module is used to acquire historical inventory information, wherein the historical inventory information includes: historical inventory data, promotion activity data of corresponding goods, and seasonal factors; The forecasting module is used to organize and clean the acquired information and predict future inventory demand through seasonal ARIMA; The prediction of future inventory demand through seasonal ARIMA includes: Divide historical data into training set and validation set; Use the training set data to train the initial demand forecast model; Use the initial demand forecasting model to forecast the validation set data, and modify the initial demand forecasting model according to the forecasting result to obtain a demand forecasting model; Input the current inventory information into the demand forecasting model to obtain the predicted inventory demand quantity Dj; The reorder point is calculated based on the predicted inventory demand quantity, and the reorder point calculation model is: Where R is the reorder point of the goods, μ is the predicted average demand quantity of the goods, LT is the average delivery time, Z is the safety factor of the goods corresponding to the required service level, n is the number of forecast points, and D i is the actual inventory demand quantity, D j To forecast inventory requirements; The order reminder module is used to send alarm information to managers to remind them to place orders when the inventory of goods reaches or falls below the reorder point; The sales allocation cargo space module is used to obtain the incoming order of the daily necessities logistics storage warehouse after placing an order, predict the sales of the goods in the incoming order, and allocate cargo space for goods with high sales volume according to the predicted sales of the goods; The module for associating and allocating cargo locations is used to obtain information about different goods being browsed, added to shopping carts, and purchased, obtain the degree of association of goods based on the information, and allocate cargo locations for goods with high association with goods with high sales volume based on the degree of association of goods; Wherein, the associated allocation cargo location module includes: The goods association calculation module is used to obtain the information of different goods being browsed, added to the shopping cart, and purchased in the logistics warehouse, and calculate the goods association through the goods association model. The cargo association model is: Among them, I(i; j t ) is the correlation between goods i and goods j at time t, x i ,x j represents the specific behaviors of goods i and goods j, including purchase, browsing and adding to shopping cart, p(x i ,x j ,t) represents the simultaneous occurrence of cargo i in state x at time t i and good j is in state x j The probability, p(x i ,t) represents that at time t, the item i is in state x i The independent probability of occurrence, p(x j ,t) represents that at time t, the commodity j is in state x j The independent probability of occurrence; A sales volume correlation module is used to obtain the correlation between the goods ranked in the bottom 50% of the predicted sales volume and the goods ranked in the top 50% of the predicted sales volume; The sorting and allocation module is used to sort the predicted sales volume of the top 50% of goods from high to low, sort the predicted sales volume of the bottom 50% of goods with respect to the top 50% of goods, obtain the goods in the bottom 50% with the greatest correlation with the goods in the top 50%, and allocate them to the vacant storage space closest to the goods in the top 50%. When the goods with the greatest correlation have been allocated a storage space, obtain the goods with the greatest correlation that have not been allocated a storage space, and allocate them to the storage space closest to the goods in the top 50%. The outbound module is used to determine the order of goods outbound according to the order picking time within a preset time interval when goods are outbound.

5. A logistics warehousing management system according to claim 4, characterized in that: The sales allocation cargo location module includes: The module for obtaining sales data of each commodity is used to obtain the warehouse entry list after placing an order, obtain the type of goods in the warehouse entry list, and obtain the historical sales data, inventory data and seasonal change data of each commodity; The sales forecasting module is used to forecast the sales of each product based on historical sales data, inventory data, and seasonal change data; Sales sorting module, used to sort goods in descending order according to predicted sales volume; The distance sorting module is used to sort the storage locations in ascending order according to the picking distance from the starting picking point to the storage location; The allocation module is used to allocate the first 50% of the goods to the sorted cargo positions in sequence.

6. A logistics warehousing management system according to claim 4, characterized in that: The outbound module includes: An order picking time acquisition module is used to acquire the picking time of each order when multiple orders are received within a preset time interval; The picking time sorting module is used to ship the orders within the preset time interval in the order of order picking time from least to most; The instruction sending module is used to send the transport instruction to the picking device, and the picking device picks the goods in sequence.

Citation Information

Patent Citations

  • Logistics center storage position allocation method

    CN106709692A

  • Commodity storage allocation method and device and storage medium

    CN107563702A

  • Warehouse storage location allocation method and system based on big data

    CN111476413A

  • Warehouse management system

    CN112407740A

  • Goods allocation two-stage optimization method and system based on improved order association rule

    CN117371918A