CN117114584B

CN117114584B - Intelligent commodity matching and compensating method

Info

Publication number: CN117114584B
Application number: CN202311387254.9A
Authority: CN
Inventors: 严嘉星; 杨涛; 万华; 廖波涛; 陆艳贞; 陈彬
Original assignee: Guangzhou Hengkang Information Technology Co ltd
Current assignee: Guangzhou Hengkang Information Technology Co ltd
Priority date: 2023-10-25
Filing date: 2023-10-25
Publication date: 2024-01-26
Anticipated expiration: 2043-10-25
Also published as: CN117114584A

Abstract

The invention provides an intelligent commodity matching and compensating method, and relates to the technical field of big data analysis. The method comprises the steps of obtaining commodity sales data of shops in an area, and carrying out sales feature analysis to form commodity sales feature data of the shops; determining the ordering amount of different commodities in each store according to the commodity sales characteristic data to form store ordering amount data; determining actual order information according to the store order quantity data, and performing first commodity allocation to form store first commodity allocation data; acquiring real-time cargo quantity information of a store, and carrying out replenishment transfer analysis by combining first cargo allocation data of the store to form replenishment transfer analysis data; and carrying out replenishment and allocation on different stores according to the replenishment and allocation analysis data. The commodity matching and adjustment method with comprehensive system and high accuracy can be formed by real-time processing and analysis of commodity data, so that the management efficiency is improved, and meanwhile, the economical efficiency of commodity sales can be further improved.

Description

Intelligent commodity matching and compensating method

Technical Field

The invention relates to the technical field of big data analysis, in particular to a method for intelligent commodity matching and compensation.

Background

In the context of digital conversion, the merchandise matching and retuning system becomes a key tool for branded retail establishments. However, there are more or less problems and disadvantages in commercial make-up systems currently available on the market. Most systems have difficulty in comprehensively and accurately monitoring and managing the whole life cycle of commodities, including links such as commodity planning, matching and adjustment management, commodity monitoring and sales prediction. Moreover, the traditional system mostly adopts a customized development architecture, can not be flexibly configured according to the actual business processes of different enterprises, and has high maintenance cost in the middle and later stages. In addition, enterprises cannot make decisions and adjustments in time due to the lack of analysis and processing of large amounts of real-time data by the system.

Therefore, the intelligent commodity matching and adjustment method is designed, a comprehensive commodity matching and adjustment mode with high accuracy can be formed by real-time processing and analysis of commodity data, management efficiency is improved, and meanwhile, the economical efficiency of commodity sales can be further improved, so that the method is a problem to be solved urgently at present.

Disclosure of Invention

The invention aims to provide an intelligent commodity matching and matching method, which is used for establishing basic characteristic data influencing the commodity quantity of a store in the next characteristic sales period by analyzing sales data in a commodity matching and matching area aiming at different characteristic factors. And the influence condition analysis of different characteristic factors on the commodity sales is carried out one by one on the basis of the basic characteristic data, and as various factors influencing the commodity sales are fully and carefully considered, more accurate and comprehensive order data can be obtained. In addition, after accurate and comprehensive ordering amount data are obtained, analysis of first allocation of a store and determination of a mode of supplementary allocation are carried out, a supplementary allocation system of the commodity is further perfected, the economical efficiency of commodity sales is greatly improved, and cost caused by excessive backlog of the commodity is effectively avoided.

In a first aspect, the present invention provides a method for intelligent commodity matching and adjustment, including obtaining commodity sales data of a store in an area, and performing sales feature analysis for different commodities on the store to form commodity sales feature data of the store; determining the ordering amount of different commodities in each store according to the commodity sales characteristic data to form store ordering amount data; determining actual order information according to the store order quantity data, and performing commodity first allocation on different stores based on the actual order information to form store first allocation object data; acquiring real-time cargo quantity information of a store, and carrying out replenishment transfer analysis by combining first cargo allocation data of the store to form replenishment transfer analysis data; and carrying out replenishment and allocation on different stores according to the replenishment and allocation analysis data.

According to the method, the sales data in the commodity distribution and compensation adjustment area are analyzed for different characteristic factors, so that basic characteristic data affecting the commodity quantity of the store in the next characteristic sales cycle are established. And the influence condition analysis of different characteristic factors on the commodity sales is carried out one by one on the basis of the basic characteristic data, and as various factors influencing the commodity sales are fully and carefully considered, more accurate and comprehensive order data can be obtained. In addition, after accurate and comprehensive ordering amount data are obtained, analysis of first allocation of a store and determination of a mode of supplementary allocation are carried out, a supplementary allocation system of the commodity is further perfected, the economical efficiency of commodity sales is greatly improved, and cost caused by excessive backlog of the commodity is effectively avoided.

As one possible implementation manner, acquiring commodity sales data of a store in an area, and performing sales feature analysis on different commodities for the store to form commodity sales feature data of the store, including: acquiring historical sales data of all commodities in different shops in a sales period to form historical sales data of single commodities; according to the historical sales data of the single product, performing characteristic analysis aiming at time-class sales characteristics affecting the sales volume of the commodity to form time-class characteristic analysis basic information; according to the historical sales data of the single product, performing characteristic analysis aiming at space sales characteristics affecting commodity sales, and forming space characteristic analysis basic information; according to the historical sales data of the single product, carrying out characteristic analysis aiming at price sales characteristics affecting commodity sales, and forming price characteristic analysis basic information; according to the historical sales data of the single product, carrying out characteristic analysis aiming at storage type sales characteristics affecting the sales volume of the commodity to form storage type characteristic analysis basic information; according to the historical sales data of the single product, carrying out characteristic analysis aiming at quality sales characteristics affecting the sales volume of the commodity to form quality characteristic analysis basic information; and combining different time type characteristic analysis basic information, space type characteristic analysis basic information, price type characteristic analysis basic information, storage type characteristic analysis basic information and quality type characteristic analysis basic information in each commodity to form commodity sales characteristic data.

In the present invention, it is an object to predict or determine the sales amount of goods in a store to provide important reference data for efficient management and make-up adjustment of sales of goods. Because of the variety of factors influencing the sales volume of the commodity, the factors mainly influencing the sales volume are classified into five categories, namely a time category, a space category, a price category, a storage category and a quality category, and of course, the characteristic category mentioned in the invention can be increased in the category due to subjective and objective reasons, and specific influencing factors in the five categories can be determined according to analysis requirements. Under the condition that the influence of sub factors in various types on sales volume is fully considered, sales volume data can be predicted and determined more accurately and comprehensively, and the accuracy of data analysis is greatly improved.

As one possible implementation manner, according to the historical sales data of the single product, performing characteristic analysis aiming at time-class sales characteristics affecting the sales amount of the commodity to form time-class characteristic analysis basic information, including: acquiring n characteristic sales cycles of different commodities in target storeThe sales data in the market are filtered to eliminate the influence of price sales characteristics on sales volume, and a characteristic sales period data set is formed >The method comprises the steps of carrying out a first treatment on the surface of the Wherein, each characteristic sales period data +.>Corresponding time period +.>The sales amount is->，/>M is->Sequential numbering of the divided time periods in the time dimension order; according to characteristic sales cycleData set A, determining the characteristic sales cycle of the commodity +.>Average sales per time period +.>The method comprises the steps of carrying out a first treatment on the surface of the Characteristic sales cycle for combining different commodities in target store>All average sales +.>Forming time class characteristic analysis basic information of different commodities.

In the invention, the characteristic analysis of the time-class sales characteristics is mainly used for determining the average sales volume condition of a store in a common sales period and reflecting the sales volume level of the whole commodity. Since discounting the commodity greatly affects the sales volume of the commodity, the influence of the price sales feature on the commodity sales volume needs to be eliminated when the time sales feature analysis is performed, so that the average sales level of the commodity in the common sales period can be reflected more truly. It can be understood that although some types of commodities have a certain sale season and a sale season due to the supply and demand characteristics of the commodities, the obtained average sales volume can reflect the sales volume of the commodities from the overall level, the average sales volume is taken as the basis of ordering, the demands of the sales volume of the commodities can be basically met in the characteristic sales period class, meanwhile, the average sales volume also provides basic data for analyzing other types of characteristics, the influence of the other types of characteristics on the sales volume of the commodities is reflected on the influence on the average sales volume, and comprehensive data reference and analysis can be performed in the subsequent process of designing the sales volume of the shops and even the specific commodities, so that the rationality and accuracy of commodity customization and allocation are greatly improved based on multi-influence factor analysis of the same basic parameters.

As one possible implementation, according to the historical sales data of the single product, space sales characteristics aiming at influencing the sales amount of the commodity are carried outCharacteristic analysis, forming basic information of spatial class characteristic analysis, comprising: setting a space analysis area, taking a target store as a position base point, taking the difference of the number of correlation stores selling the same target commodity as the target store as an extraction target, extracting characteristic sales periods with different correlation stores from n characteristic sales periods, and determining the characteristic sales periods as space analysis periods; grouping the space analysis periods according to the sequence from the near to the far of the time dimension to form space analysis period groups, ensuring that the number of the space analysis periods contained in each group is 3, and removing the rest space analysis periods if the number of the last rest space analysis periods is less than 3; for each space analysis period, establishing a two-dimensional position coordinate system by taking the target store as an origin, and determining the distance between the correlation store and the target storeU is the number of the corresponding correlation store and determines the relative space rateThe method comprises the steps of carrying out a first treatment on the surface of the Determining the sales amount of each correlation store in the corresponding spatial analysis period>And determining the average sales amount +. >The method comprises the steps of carrying out a first treatment on the surface of the Determining target sales amount of target commodity of target store in corresponding space analysis periodThe method comprises the steps of carrying out a first treatment on the surface of the According to the relative spatial rate->Average sales amount->Total sales target->And the number u of correlation stores, determining the spatial divisionSpace constraint parameter corresponding to analysis period>Wherein->The method comprises the steps of carrying out a first treatment on the surface of the Three space constraint parameters in each space analysis period group are acquired, an objective function of h is established, and the relative space rate is increased>And the number u of correlation stores is a binary function of the variables; homogenizing according to binary function of each space analysis period group to form space restriction function +.>The method comprises the steps of carrying out a first treatment on the surface of the And combining space constraint functions of all commodities in the target store to form space class characteristic analysis basic information.

In the invention, the analysis of the space sales characteristics is performed for the commodity sales volume mainly by considering the mutual restriction of sales volume generated by the distance of the stores selling the same commodity due to the geographical position, namely, the stores selling the same commodity compete for the demand clients in a certain area, thereby leading to competition in commodity sales volume. While the effect of the space-like sales feature on sales is mainly in three aspects, one is the number of stores selling the same commodity, the more the number of stores selling the same commodity near the target store, the greater the commodity sales pressure of the target store, and the second is the distance position of the store selling the same commodity from the target store with respect to the target store, the closer the influence from the target store is. The third aspect is the target commodity sales amount of the correlation store, and the larger the target commodity sales amount of the correlation store, the larger the target commodity sales amount of the target store will fluctuate. Therefore, the influence of the store selling the same commodity on the commodity sales of the target store can be determined by establishing a functional relation between the target commodity sales of the target store and the position parameters and the number of the correlated commodities, and the main factors which affect the target commodity sales of the target store after all are the number and the position parameters of the correlated stores, considering that the store selling the same commodity is basically unchanged in the space analysis area when the data are extracted from n characteristic sales periods and the targeted commodity types are only adjusted after the sale for a long time. In addition, as for the space constraint function, it is essential to give a rate of change of the space class feature to the sales amount of the commodity sold by the store, that is, a rate at which the average sales amount of the target commodity of the target store increases or decreases relative to the average sales amount in the case of the relevant store. Further, the grouping process of the space analysis period is started by the latest feature sales period, because the data closer to the time has more referential, the triad is used for determining the coefficient of the binary function, and the unit of one group is used for carrying out the homogenization treatment of the binary function, wherein the homogenization treatment can be used for establishing function images through the binary functions of different groups and carrying out homogenization on the results under the same variable to form a new binary function.

As one possible implementation manner, according to the historical sales data of the single product, performing characteristic analysis of price type sales characteristics affecting sales volume of the commodity to form price type characteristic analysis basic information, including: determining a price analysis area, and acquiring unified discount values of target commodities of different shops in n characteristic sales periods in the price analysis areaUniform discount value->Corresponding unified discount period->Uniform discount sales->V is the number corresponding to the store selling the target commodity in the characteristic sales period with the number x, and x represents the extracted uniform fold in the n characteristic sales periodsThe buckling period is->A number of characteristic sales cycles of (a); determining and unifying discount periods according to time class characteristic analysis basic informationAdjacent and in the unified discounts period->Average sales in the preceding time period +.>And unify discounts periodsAdjacent and in the unified discounts period->Average sales>The method comprises the steps of carrying out a first treatment on the surface of the Determining an average unified discount rate of change according to>：/>，/>Wherein->Obtaining in-store discount values for different shops for target goods in n characteristic sales periods +.>In-store discount value->Corresponding in-store discount period->In-store discount sales->I represents the number of the store selling the target commodity, and y represents the number of the characteristic selling period of the in-store discount performed by the store with the number i; analyzing basic information according to time-class characteristics, and determining the discount period of different shops in the shop >Corresponding characteristic sales period in-store discount period +.>Adjacent and in-store discount period->Average sales in the preceding time period +.>Adjacent and in-store discount periodsAverage sales>The method comprises the steps of carrying out a first treatment on the surface of the Determining average in-store discount rate of change for different stores according to>：，/>，/>The method comprises the steps of carrying out a first treatment on the surface of the Combined average unified discount rate of change->And all average in-store discount rate of change +.>And forming price class characteristic analysis basic information.

In the present invention, the impact of the price class sales feature on the sales of the merchandise is significant. The analysis of the price class sales feature here includes two, one is the effect of the unified discount value set by the commodity itself on the sales volume, and the other is the effect of the in-store discount value given by the individual store for its own situation on the sales volume. The influence of the unified discount value on the commodity sales volume is based on analysis of target commodities sold by all shops in a price analysis area, namely, discount sales volume of different shops in each characteristic sales period in the unified discount value period is extracted from n characteristic sales periods, and meanwhile, the change rate of the sales volume change volume relative to the unified discount period time is obtained by taking the average sales volume of the discount sales volume relative to the common sales period of the commodities in the shops as a reference to represent the increase condition of the unified discount value on the average commodity sales volume. Since some kinds of commodities are sold in off-season and in high-season, the average sales amount of the common sales period before and after the uniform discount period is used as a reference when considering the average sales amount of the discount sales amount relative to the common sales period of the commodities in the store, and thus the analysis of the variation amount is more accurate. Of course, since the price type sales feature has a requirement for the time period, the time period division can be performed with reference to the discount period analyzed according to the price type sales feature. As for the influence of the in-store discount value on the sales of the commodity, since the discount value is different for each store, the analysis is performed in units of different stores, and finally, the influence result of the in-store discount on the sales of the same commodity of different stores is subjected to homogenization processing. Here, considering that the discount values are different from store to store, when the influence of the discount value on the sales volume of the commodity is carried out, the influence of the discount value on the sales volume is converted into the influence of the difference value of the discount value in the store relative to the uniform discount value on the sales volume, so that a uniform reference basis is provided when data processing is carried out, and the analysis of the discount value in the store relative to the uniform discount value is also facilitated, and the sales volume of the commodity is facilitated. Also, since there are some kinds of products in which there are off-season and on-season sales for the influence of the in-store discount value on the sales volume of the products, when considering the average sales volume of the discount sales volume relative to the average sales volume of the products in the ordinary sales period in the store, the average sales volume of the ordinary sales period before and after the in-store discount period is taken as a reference, and thus the analysis of the variation volume is more accurate.

As one possible implementation manner, according to the historical sales data of the single product, performing characteristic analysis of storage type sales characteristics affecting sales volume of the commodity to form storage type characteristic analysis basic information, including: acquiring remaining stock amounts of target commodities in different stores in n characteristic sales periodsAnd corresponding sales ∈ ->Z represents the store number of the sales target commodity, k is the sequence number of the characteristic sales cycle of the sales target commodity extracted in n characteristic sales cycles in the time dimension; according to the remaining stock quantity->And sales +.>Determining the pin memory ratio->Wherein:the method comprises the steps of carrying out a first treatment on the surface of the According to the pin memory ratio->Determining a pin ratio difference between adjacent characteristic sales cycles in a time dimension +.>，2≤/>K is less than or equal to; according to the pin ratio differenceEstablishing a time-dimension-based function of the rate of change of sales of the target commodity in different stores +.>The method comprises the steps of carrying out a first treatment on the surface of the Combine all pin memory rate functions->Storage class characteristic analysis base information is formed.

In the invention, the analysis of the influence of the storage sales characteristics on the commodity sales is to determine the change trend of the ratio of the residual inventory quantity to the sales quantity of the commodities in the store, so that reasonable prediction is conveniently provided when the commodity ordering quantity is determined for the next characteristic sales period, excessive commodity surplus and extrusion are avoided, and the cost is increased while the inventory resource is occupied.

As one possible implementation manner, according to the historical sales data of the single product, performing characteristic analysis aiming at quality type sales characteristics affecting sales volume of the commodity to form quality type characteristic analysis basic information, including: analyzing basic information according to time characteristics to obtain sales of target commodities in each of n characteristic sales periodsAnd sales amount of the same type of goods as the target goods +.>E denotes the number of the same type of commodity as the target commodity, wherein>，/>The method comprises the steps of carrying out a first treatment on the surface of the In each characteristic sales period, obtaining the sales difference between the similar goods and the target goods>And according to the sales difference, establishing a sales difference change rate function of the target commodity relative to the similar commodity based on the time dimension +.>The method comprises the steps of carrying out a first treatment on the surface of the All sales difference change rate functions of combined target commodity relative to similar commodity +.>And forming quality class characteristic analysis basic information.

In the invention, the characteristic analysis of the quality sales characteristic is to compare sales of the same type of commodities so as to eliminate the similar commodities without sales potential and avoid cost increase. The sales quantity comparison of similar commodities is characterized by a change function of sales difference distance along with time, and the sales quantity of the similar commodities in the next characteristic sales period can be predicted according to the sales difference change rate function, so that the commodity types of ordered commodities are optimized, and the sales quantity is improved.

As one possible implementation, determining an order amount of different commodities for each store according to commodity sales feature data, forming store order amount data, includes: determining the goods s sold in the store and determining the order amount of each of the goods in the store according to the following analysis and judgment stepsS is the number of all different kinds of commodities: setting a sales difference change rate threshold +.>According to the sales difference rate function ∈ ->Determining the sales difference change rate of the product s relative to the similar products in the next characteristic sales cycle +.>And the following judgment is made: if->The store no longer supplies items s for the next characteristic sales cycle; if->The store continues to supply the goods s in the next characteristic sales cycle and determines the order quantity of the goods s in accordance with the following manner>: analyzing the basic information according to the time-class characteristics to determine the total average sales of the commodity in the characteristic sales period>Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the Acquiring the number u of stores selling the same items s as the stores in the next characteristic sales period, the relative space rate +.>According to the space constraint function->Determining the space restriction rate of commodity s>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring a preset unified discount period of the store in the next characteristic sales cycle >And preset unified discount value->And unifies the discount rate of change according to the average of the goods s +.>Determining whether the commodity s is in a preset unified discount period>Sales amount of->Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the Acquiring a preset in-store discount period of a store in a next characteristic sales period>And preset in-store discount value->And is based on the preset in-store discount period of commodity s +.>Preset in-store discount valuePresetting unified discount value->Determining the discount period of the commodity s in the preset store>Sales amount of->Wherein:at this time i is the preset in-store discount period +.>Corresponding store numbers; setting a pin ratio threshold->According to the pin memory rate function->Determining the stock-out ratio of the item s in the next characteristic sales cycle in the store +.>And makes the following decisions: if->Then use the threshold value of the stock ratio +.>As the stock/sales ratio of the article s in the next characteristic sales period, and determining the remaining storage amount +.>Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the If->According to->Determining remaining storage of item s in the next characteristic sales cycle +.>Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the According to the total average sales of goods s +.>Space restriction rate->Presetting unified discount period->Sales amount of->Preset in-store discount period- >Sales amount of->Residual memory +.>Determining the order amount of the goods s in the store +.>Wherein:the method comprises the steps of carrying out a first treatment on the surface of the Order amount according to different goods s in store +.>Store order data is determined.

In the present invention, commodity order amount data is completed by comprehensively considering time-type characteristic analysis base information, space-type characteristic analysis base information, price-type characteristic analysis base information, storage-type characteristic analysis base information, and quality-type characteristic analysis base information. Firstly, based on quality characteristic analysis basic information, judging whether the commodity is required to be ordered, and then taking the total average sales volume as a reference to respectively obtain space constraint rateAnd integrating the unified discount sales, the in-store discount sales and the rest storage classes to finally determine the ordering amount of the ordered goods. The influence of a plurality of factors on the ordering amount is comprehensively considered, so that the ordering amount is more reasonable and accurate, and the determination mode based on big data is also efficient and quick.

As one possible implementation manner, determining actual order information according to store order quantity data, and performing first distribution of goods to different stores based on the actual order information to form first distribution data of the stores, including: acquiring sales task amount of store in next characteristic sales period and ordering amount according to commodity s Determining additional turnover of a storeThe method comprises the steps of carrying out a first treatment on the surface of the Determining store first shipment data based on the order volume and additional turnover volume for different items within the store, wherein: acquiring sales task amount of store in next characteristic sales period and ordering goods according to goods s>Determining additional turnover of store +.>Comprising: acquiring the sales task amount of the store in the next characteristic sales period, and determining the residual memory capacity of the store by combining the ordering amount of different commodities in the store; determining the total ordering amount of the same commodity s in the next characteristic sales period according to the ordering amount of the same commodity s in different stores, sorting the commodity s from large to small according to the total ordering amount, extracting the total ordering amount of the previous p commodities, and determining the total ordering amount ratio of the p commodities; and acquiring the storage amounts of all stores, and respectively giving the storage amounts to p commodities according to the total order quantity ratio.

In the invention, since a certain sales index exists in the store, after the commodity ordering amount of the store is determined, the difference existing based on the sales index can be determined by supplementing the hot-sold commodity, so that the sales index can be completed, and the sales amount of the supplemented commodity can be ensured, thereby avoiding the increase of the cost.

As one possible implementation manner, acquiring real-time cargo amount information of a store, and performing replenishment transfer analysis in combination with first-order cargo amount data of the store to form replenishment transfer analysis data, including: setting a stock early warning lower limit valueAnd stock early warning upper limit +.>And->The method comprises the steps of carrying out a first treatment on the surface of the Acquiring real-time stock of all commodities in a store, and when the real-time stock of the commodities is not more than the stock early warning lower limit value +.>When the stores are used as position points, the stores in other stores are sequentially determined to exceed the stock early warning upper limit value according to the sequence from near to far of the relative distance>And extracting the upper limit value of the stock quantity early warning +.>Until the accumulated and extracted stock guarantees that the stock reaches the stock early warning upper limit value after the store is supplemented +.>Until that point.

In the invention, the replenishment and allocation of the commodities are carried out by taking the backout store as a reference and taking the distance and the inventory of the same commodities in other stores as indexes, so that the replenishment and allocation can be carried out to the target store in the fastest mode while the sales of the same commodities in other stores is ensured to obtain inventory assurance, and the replenishment and allocation are efficient and quick.

The intelligent commodity matching and supplementing method provided by the invention has the beneficial effects that:

According to the method, the sales data in the commodity matching and adjustment area are analyzed aiming at different characteristic factors, so that basic characteristic data influencing the commodity quantity of the store in the next characteristic sales period are established. And the influence condition analysis of different characteristic factors on the commodity sales is carried out one by one on the basis of the basic characteristic data, and as various factors influencing the commodity sales are fully and carefully considered, more accurate and comprehensive order data can be obtained. In addition, after accurate and comprehensive ordering amount data are obtained, analysis of first allocation of a store and determination of a mode of supplementary allocation are carried out, a supplementary allocation system of the commodity is further perfected, the economical efficiency of commodity sales is greatly improved, and cost caused by excessive backlog of the commodity is effectively avoided.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments of the present invention will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and should not be considered as limiting the scope, and other related drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a step diagram of a method for intelligent commodity matching and adjustment according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.

Referring to fig. 1, an embodiment of the present invention provides a method for intelligent commodity matching and adjustment. According to the method, the sales data in the commodity matching and adjustment area are analyzed aiming at different characteristic factors, so that basic characteristic data influencing the commodity quantity of the store in the next characteristic sales period are established. And the influence condition analysis of different characteristic factors on the commodity sales is carried out one by one on the basis of the basic characteristic data, and as various factors influencing the commodity sales are fully and carefully considered, more accurate and comprehensive order data can be obtained. In addition, after accurate and comprehensive ordering amount data are obtained, analysis of first allocation of a store and determination of a mode of supplementary allocation are carried out, a supplementary allocation system of the commodity is further perfected, the economical efficiency of commodity sales is greatly improved, and cost caused by excessive backlog of the commodity is effectively avoided.

The intelligent commodity matching and supplementing method specifically comprises the following steps:

s1: and acquiring commodity sales data of the shops in the area, and analyzing sales characteristics of the shops for different commodities to form commodity sales characteristic data of the shops.

Acquiring commodity sales data of a store in an area, and analyzing sales characteristics of different commodities of the store to form commodity sales characteristic data of the store, wherein the commodity sales characteristic data comprises: acquiring historical sales data of all commodities in different shops in a sales period to form historical sales data of single commodities; according to the historical sales data of the single product, performing characteristic analysis aiming at time-class sales characteristics affecting the sales volume of the commodity to form time-class characteristic analysis basic information; according to the historical sales data of the single product, performing characteristic analysis aiming at space sales characteristics affecting commodity sales, and forming space characteristic analysis basic information; according to the historical sales data of the single product, carrying out characteristic analysis aiming at price sales characteristics affecting commodity sales, and forming price characteristic analysis basic information; according to the historical sales data of the single product, carrying out characteristic analysis aiming at storage type sales characteristics affecting the sales volume of the commodity to form storage type characteristic analysis basic information; according to the historical sales data of the single product, carrying out characteristic analysis aiming at quality sales characteristics affecting the sales volume of the commodity to form quality characteristic analysis basic information; and combining different time type characteristic analysis basic information, space type characteristic analysis basic information, price type characteristic analysis basic information, storage type characteristic analysis basic information and quality type characteristic analysis basic information in each commodity to form commodity sales characteristic data.

The method aims at predicting or determining the sales amount of the commodity in the store so as to provide important reference data for efficient management and matching and adjustment of commodity sales. Because of the variety of factors influencing the sales volume of the commodity, the factors mainly influencing the sales volume are classified into five categories, namely a time category, a space category, a price category, a storage category and a quality category, and of course, the characteristic category mentioned in the invention can be increased in the category due to subjective and objective reasons, and specific influencing factors in the five categories can be determined according to analysis requirements. Under the condition that the influence of sub factors in various types on sales volume is fully considered, sales volume data can be predicted and determined more accurately and comprehensively, and the accuracy of data analysis is greatly improved.

According to the historical sales data of the single product, characteristic analysis aiming at time-class sales characteristics affecting commodity sales is carried out to form time-class characteristic analysis basic information, and the method comprises the following steps: acquiring n characteristic sales cycles of different commodities in target storeThe sales data in the market are filtered to eliminate the influence of price sales characteristics on sales volume, and a characteristic sales period data set is formed >The method comprises the steps of carrying out a first treatment on the surface of the Wherein, each characteristic sales period data +.>Corresponding time period +.>The sales amount is->，/>m is->Sequential numbering of the divided time periods in the time dimension order; determining the characteristic sales period of the commodity according to the characteristic sales period data set A>Average sales per time period +.>The method comprises the steps of carrying out a first treatment on the surface of the Characteristic sales cycle for combining different commodities in target store>All average sales +.>Forming time class characteristic analysis basic information of different commodities.

The characteristic analysis of the time-class sales characteristics is mainly to determine the average sales volume condition of the store in the ordinary sales period and is used for reflecting the sales volume level of the whole commodity. Since discounting the commodity greatly affects the sales volume of the commodity, the influence of the price sales feature on the commodity sales volume needs to be eliminated when the time sales feature analysis is performed, so that the average sales level of the commodity in the common sales period can be reflected more truly. It can be understood that although some types of commodities have a certain sale season and a sale season due to the supply and demand characteristics of the commodities, the obtained average sales volume can reflect the sales volume of the commodities from the overall level, the average sales volume is taken as the basis of ordering, the demands of the sales volume of the commodities can be basically met in the characteristic sales period class, meanwhile, the average sales volume also provides basic data for analyzing other types of characteristics, the influence of the other types of characteristics on the sales volume of the commodities is reflected on the influence on the average sales volume, and comprehensive data reference and analysis can be performed in the subsequent process of designing the sales volume of the shops and even the specific commodities, so that the rationality and accuracy of commodity customization and allocation are greatly improved based on multi-influence factor analysis of the same basic parameters.

According to the historical sales data of the single product, carrying out characteristic analysis aiming at space sales characteristics affecting commodity sales, forming space characteristic analysis basic information, and comprising the following steps: setting a space analysis area, taking a target store as a position base point to be the same as the sales of the target storeThe number of the correlation stores of the target commodity is different as an extraction target, and the characteristic sales periods with different numbers of the correlation stores are extracted from n characteristic sales periods and are determined as space analysis periods; grouping the space analysis periods according to the sequence from the near to the far of the time dimension to form space analysis period groups, ensuring that the number of the space analysis periods contained in each group is 3, and removing the rest space analysis periods if the number of the last rest space analysis periods is less than 3; for each space analysis period, establishing a two-dimensional position coordinate system by taking the target store as an origin, and determining the distance between the correlation store and the target storeU is the number of the corresponding correlation store and determines the relative spatial rate +.>The method comprises the steps of carrying out a first treatment on the surface of the Determining the sales amount of each correlation store in the corresponding spatial analysis period>And determining the average sales amount +.>The method comprises the steps of carrying out a first treatment on the surface of the Determining a target sales amount +.A target commodity of a target store in a corresponding spatial analysis period >The method comprises the steps of carrying out a first treatment on the surface of the According to the relative spatial rate->Average sales amount->Total sales target->And the number u of correlation stores, determining a spatial constraint parameter corresponding to the spatial analysis period +.>Wherein->The method comprises the steps of carrying out a first treatment on the surface of the Three space constraint parameters in each space analysis period group are acquired, an objective function of h is established, and the relative space rate is increased>And the number u of correlation stores is a binary function of the variables; homogenizing according to binary function of each space analysis period group to form space restriction function +.>The method comprises the steps of carrying out a first treatment on the surface of the And combining space constraint functions of all commodities in the target store to form space class characteristic analysis basic information.

The analysis of the space sales characteristic aiming at the commodity sales mainly considers the mutual restriction on sales volume generated by the distance of the stores selling the same commodity due to the geographical position, namely, the stores selling the same commodity compete for the demand clients in a certain area, thereby leading to competition in commodity sales volume. While the effect of the space-like sales feature on sales is mainly in three aspects, one is the number of stores selling the same commodity, the more the number of stores selling the same commodity near the target store, the greater the commodity sales pressure of the target store, and the second is the distance position of the store selling the same commodity from the target store with respect to the target store, the closer the influence from the target store is. The third aspect is the target commodity sales amount of the correlation store, and the larger the target commodity sales amount of the correlation store, the larger the target commodity sales amount of the target store will fluctuate. Therefore, the influence of the store selling the same commodity on the commodity sales of the target store can be determined by establishing a functional relation between the target commodity sales of the target store and the position parameters and the number of the correlated commodities, and the main factors which affect the target commodity sales of the target store after all are the number and the position parameters of the correlated stores, considering that the store selling the same commodity is basically unchanged in the space analysis area when the data are extracted from n characteristic sales periods and the targeted commodity types are only adjusted after the sale for a long time. In addition, as for the space constraint function, it is essential to give a rate of change of the space class feature to the sales amount of the commodity sold by the store, that is, a rate at which the average sales amount of the target commodity of the target store increases or decreases relative to the average sales amount in the case of the relevant store. Further, the grouping process of the space analysis period is started by the latest feature sales period, because the data closer to the time has more referential, the triad is used for determining the coefficient of the binary function, and the unit of one group is used for carrying out the homogenization treatment of the binary function, wherein the homogenization treatment can be used for establishing function images through the binary functions of different groups and carrying out homogenization on the results under the same variable to form a new binary function.

According to the historical sales data of the single product, carrying out characteristic analysis aiming at price type sales characteristics influencing commodity sales, and forming price type characteristic analysis basic information, wherein the method comprises the following steps: determining a price analysis area, and acquiring unified discount values of target commodities of different shops in n characteristic sales periods in the price analysis areaUniform discount value->Corresponding unified discount period->Uniform discount sales->V is the number corresponding to the store selling the target commodity in the characteristic sales period of number x, x represents the extracted ++a ∈with uniform discount period in the n characteristic sales periods>A number of characteristic sales cycles of (a); according to time-like characteristicsAnalyzing the basic information to determine and unify the discount period->Adjacent and in the unified discounts period->Average sales in the preceding time period +.>And unify discounts period->Adjacent and during a unified discount periodAverage sales>The method comprises the steps of carrying out a first treatment on the surface of the Determining an average unified discount rate of change according to>：，/>Wherein->The method comprises the steps of carrying out a first treatment on the surface of the Obtaining in-store discount values for different shops for target goods in n characteristic sales periods +.>In-store discount value->Corresponding in-store discount period->In-store discount sales->I represents the number of the store selling the target commodity, and y represents the number of the characteristic selling period of the in-store discount performed by the store with the number i; analyzing basic information according to time-class characteristics, and determining the discount period of different shops in the shop >Corresponding characteristic sales period in-store discount period +.>Adjacent and in-store discount period->Average sales in the preceding time period +.>And +.>Corresponding characteristic sales period in-store discount period +.>Adjacent and in-store discount period->Average sales afterThe method comprises the steps of carrying out a first treatment on the surface of the Determining average in-store discount rate of change for different stores according to>：/>，，/>The method comprises the steps of carrying out a first treatment on the surface of the Combining average unified discounted rate of changeAnd all average in-store discount rate of change +.>And forming price class characteristic analysis basic information.

The impact of price class sales features on the sales of goods is significant. The analysis of the price class sales feature here includes two, one is the effect of the unified discount value set by the commodity itself on the sales volume, and the other is the effect of the in-store discount value given by the individual store for its own situation on the sales volume. The influence of the unified discount value on the commodity sales volume is based on analysis of target commodities sold by all shops in a price analysis area, namely, discount sales volume of different shops in each characteristic sales period in the unified discount value period is extracted from n characteristic sales periods, and meanwhile, the change rate of the sales volume change volume relative to the unified discount period time is obtained by taking the average sales volume of the discount sales volume relative to the common sales period of the commodities in the shops as a reference to represent the increase condition of the unified discount value on the average commodity sales volume. Since some kinds of commodities are sold in off-season and in high-season, the average sales amount of the common sales period before and after the uniform discount period is used as a reference when considering the average sales amount of the discount sales amount relative to the common sales period of the commodities in the store, and thus the analysis of the variation amount is more accurate. Of course, since the price type sales feature has a requirement for the time period, the time period division can be performed with reference to the discount period analyzed according to the price type sales feature. As for the influence of the in-store discount value on the sales of the commodity, since the discount value is different for each store, the analysis is performed in units of different stores, and finally, the influence result of the in-store discount on the sales of the same commodity of different stores is subjected to homogenization processing. Here, considering that the discount values are different from store to store, when the influence of the discount value on the sales volume of the commodity is carried out, the influence of the discount value on the sales volume is converted into the influence of the difference value of the discount value in the store relative to the uniform discount value on the sales volume, so that a uniform reference basis is provided when data processing is carried out, and the analysis of the discount value in the store relative to the uniform discount value is also facilitated, and the sales volume of the commodity is facilitated. Also, since there are some kinds of products in which there are off-season and on-season sales for the influence of the in-store discount value on the sales volume of the products, when considering the average sales volume of the discount sales volume relative to the average sales volume of the products in the ordinary sales period in the store, the average sales volume of the ordinary sales period before and after the in-store discount period is taken as a reference, and thus the analysis of the variation volume is more accurate.

According to the historical sales data of the single product, carrying out characteristic analysis aiming at storage type sales characteristics affecting commodity sales, forming storage type characteristic analysis basic information, and comprising the following steps: acquiring remaining stock amounts of target commodities in different stores in n characteristic sales periodsAnd corresponding sales ∈ ->Z represents the store number of the sales target commodity, k is the sequence number of the characteristic sales cycle of the sales target commodity extracted in n characteristic sales cycles in the time dimension; according to the remaining stock quantity->And sales +.>Determining the pin memory ratio->Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the According to the pin memory ratio->Determining pin ratio differences between adjacent feature sales cycles in a time dimension，/>The method comprises the steps of carrying out a first treatment on the surface of the According to the pin ratio difference->Establishing a time-dimension-based function of the rate of change of sales of the target commodity in different stores +.>The method comprises the steps of carrying out a first treatment on the surface of the Combine all pin memory rate functions->Storage class characteristic analysis base information is formed.

The analysis of the influence of the storage type sales feature on the commodity sales is to determine the change trend of the ratio of the residual inventory quantity to the sales quantity of the commodity in the store, so that reasonable prediction is conveniently provided when the commodity ordering quantity is determined for the next feature sales period, excessive commodity surplus and extrusion are avoided, and the cost is increased while the inventory resource is occupied.

According to the historical sales data of the single product, carrying out characteristic analysis aiming at quality type sales characteristics affecting commodity sales, forming quality type characteristic analysis basic information, and comprising the following steps: analyzing basic information according to time characteristics to obtain sales of target commodities in each of n characteristic sales periodsAnd sales amount of the same type of goods as the target goods +.>E denotes the number of the same type of commodity as the target commodity, wherein,，/>the method comprises the steps of carrying out a first treatment on the surface of the In each characteristic sales period, similar commodities and targets are obtainedSales of goods are bad->And according to the sales difference, establishing a sales difference change rate function of the target commodity relative to the similar commodity based on the time dimension +.>The method comprises the steps of carrying out a first treatment on the surface of the All sales difference change rate functions of combined target commodity relative to similar commodity +.>And forming quality class characteristic analysis basic information.

The characteristic analysis of the quality sales characteristics is to compare sales of the same type of commodities so as to eliminate the similar commodities without sales potential and avoid cost increase. The sales quantity comparison of similar commodities is characterized by a change function of sales difference distance along with time, and the sales quantity of the similar commodities in the next characteristic sales period can be predicted according to the sales difference change rate function, so that the commodity types of ordered commodities are optimized, and the sales quantity is improved.

S2: and determining the ordering amount of different commodities for each store according to the commodity sales characteristic data, and forming store ordering amount data.

Determining the ordering amount of different commodities for each store according to commodity sales characteristic data to form store ordering amount data, comprising: determining the goods s sold in the store and determining the order amount of each of the goods in the store according to the following analysis and judgment stepsS is the number of all different kinds of commodities: setting a sales difference change rate threshold +.>According to the sales difference rate function ∈ ->Determining sales variation of the commodity s relative to the similar commodity in the next characteristic sales periodTransformation rate->And makes the following decisions: if->The store no longer supplies items s for the next characteristic sales cycle; if it isThe store continues to supply the goods s in the next characteristic sales cycle and determines the order quantity of the goods s in accordance with the following manner>: analyzing the basic information according to the time-class characteristics to determine the total average sales of the commodity in the characteristic sales period>Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the Acquiring the number u of stores selling the same items s as the stores in the next characteristic sales period, the relative space rate +.>According to the space constraint function- >Determining the space restriction rate of commodity s>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring a preset unified discount period of the store in the next characteristic sales cycle>And preset unified discount value->And unifies the discount rate of change according to the average of the goods s +.>Determining whether the commodity s is in a preset unified discount period>Sales amount of->Wherein:the method comprises the steps of carrying out a first treatment on the surface of the Acquiring a preset in-store discount period of a store in a next characteristic sales period>And preset in-store discount value->And is based on the preset in-store discount period of commodity s +.>Preset in-store discount value->Presetting unified discount value->Determining the discount period of the commodity s in the preset store>Sales amount of->Wherein:at this time i is the preset in-store discount period +.>Corresponding store numbers; setting a pin ratio threshold->According to the pin memory rate function->Determining the stock-out ratio of the item s in the next characteristic sales cycle in the store +.>And makes the following decisions: if->Then use the threshold value of the stock ratio +.>As the stock/sales ratio of the article s in the next characteristic sales period, and determining the remaining storage amount +.>Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the If->According to->Determining remaining storage of item s in the next characteristic sales cycle +.>Wherein: />The method comprises the steps of carrying out a first treatment on the surface of the According to the total average sales of goods s +. >Space restriction rate->Presetting unified discount period->Sales amount of->Preset in-store discount period->Sales amount of->Residual memory +.>Determining the order amount of the goods s in the store +.>Wherein:the method comprises the steps of carrying out a first treatment on the surface of the Order amount according to different goods s in store +.>Store order data is determined.

The commodity order quantity data is completed by comprehensively considering the time type characteristic analysis basic information, the space type characteristic analysis basic information, the price type characteristic analysis basic information, the storage type characteristic analysis basic information and the quality type characteristic analysis basic information. Firstly, based on quality characteristic analysis basic information, judging whether the commodity is required to be ordered, and then taking the total average sales volume as a reference to respectively obtain space constraint rateAnd integrating the unified discount sales, the in-store discount sales and the rest storage classes to finally determine the ordering amount of the ordered goods. The influence of a plurality of factors on the ordering amount is comprehensively considered, so that the ordering amount is more reasonable and accurate, and the determination mode based on big data is also efficient and quick.

S3: and determining actual order information according to the store order quantity data, and performing commodity first allocation on different stores based on the actual order information to form store first allocation object data.

According to shopsThe order quantity data is used for determining actual order information, and carrying out commodity first allocation on different shops based on the actual order information to form shop first allocation object data, and the method comprises the following steps: acquiring sales task amount of store in next characteristic sales period and ordering amount according to commodity sDetermining additional turnover of store +.>The method comprises the steps of carrying out a first treatment on the surface of the Determining store first shipment data based on the order volume and additional turnover volume for different items within the store, wherein: acquiring sales task amount of store in next characteristic sales period and ordering goods according to goods s>Determining additional turnover of store +.>Comprising: acquiring the sales task amount of the store in the next characteristic sales period, and determining the residual memory capacity of the store by combining the ordering amount of different commodities in the store; determining the total ordering amount of the same commodity s in the next characteristic sales period according to the ordering amount of the same commodity s in different stores, sorting the commodity s from large to small according to the total ordering amount, extracting the total ordering amount of the previous p commodities, and determining the total ordering amount ratio of the p commodities; and acquiring the storage amounts of all stores, and respectively giving the storage amounts to p commodities according to the total order quantity ratio.

Since a certain sales index exists in the store, after the commodity ordering amount of the store is determined, the difference existing based on the sales index can be determined by supplementing the hot-sold commodity, so that the sales index can be completed, the sales amount of the supplemented commodity can be ensured, and the increase of cost is avoided.

S4: and acquiring real-time cargo quantity information of the store, and carrying out replenishment transfer analysis by combining first cargo distribution data of the store to form replenishment transfer analysis data.

Acquiring real-time goods of storeThe article quantity information is combined with first article distribution data of a store to carry out replenishment transfer analysis, and replenishment transfer analysis data is formed, and the method comprises the following steps: setting a stock early warning lower limit valueAnd stock early warning upper limit +.>And->The method comprises the steps of carrying out a first treatment on the surface of the Acquiring real-time stock of all commodities in a store, and when the real-time stock of the commodities is not more than the stock early warning lower limit value +.>When the stores are used as position points, the stores in other stores are sequentially determined to exceed the stock early warning upper limit value according to the sequence from near to far of the relative distance>And extracting the upper limit value of the stock quantity early warning +.>Until the accumulated and extracted stock guarantees that the stock reaches the stock early warning upper limit value after the store is replenished Until that point.

The replenishment and allocation of the commodities are carried out by taking the backout store as a reference and taking the stock quantity of the same commodities in the distance and other stores as indexes, so that the replenishment and allocation can be carried out to the target store in the fastest mode while the sales of the same commodities in other stores is ensured to be ensured.

S5: and carrying out replenishment and allocation on different stores according to the replenishment and allocation analysis data.

And the replenishment transfer analysis data is obtained based on multi-factor sales volume analysis, so that the replenishment transfer can be rapidly carried out on different shops, and the sales volume of the shops is ensured.

In summary, the method for matching and supplementing intelligent commodity provided by the embodiment of the invention has the beneficial effects that:

In the present invention, "at least one" means one or more, and "a plurality" means two or more. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b, or c may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or plural.

It should be understood that, in various embodiments of the present invention, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

Those of ordinary skill in the art will appreciate that the elements and method steps of the examples described in connection with the embodiments disclosed herein can be implemented as electronic hardware, or as a combination of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.

In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Claims

1. The intelligent commodity matching and complement adjustment method is characterized by comprising the following steps:

acquiring commodity sales data of shops in the area, and analyzing sales characteristics of shops aiming at different commodities to form commodity sales characteristic data of shops;

determining the ordering amount of different commodities in each store according to the commodity sales characteristic data to form store ordering amount data;

determining actual order information according to the store order quantity data, and performing commodity first allocation on different stores based on the actual order information to form store first allocation object data;

acquiring real-time cargo quantity information of a store, and carrying out replenishment transfer analysis by combining first cargo data of the store to form replenishment transfer analysis data;

According to the replenishment transfer analysis data, replenishment transfer is carried out on different shops;

the method for obtaining the commodity sales data of the shops in the area, analyzing the sales characteristics of the shops aiming at different commodities, forming the commodity sales characteristic data of the shops comprises the following steps:

acquiring historical sales data of all commodities in different shops in a sales period to form historical sales data of single commodities; according to the historical sales data of the single product, performing characteristic analysis aiming at time-class sales characteristics affecting commodity sales, and forming time-class characteristic analysis basic information; according to the historical sales data of the single product, performing characteristic analysis aiming at space sales characteristics affecting commodity sales, and forming space characteristic analysis basic information; according to the historical sales data of the single product, carrying out characteristic analysis aiming at price sales characteristics affecting commodity sales, and forming price characteristic analysis basic information; according to the historical sales data of the single product, performing characteristic analysis aiming at storage type sales characteristics affecting commodity sales, and forming storage type characteristic analysis basic information; according to the historical sales data of the single product, performing characteristic analysis aiming at quality type sales characteristics affecting commodity sales, and forming quality type characteristic analysis basic information;

Combining different time type characteristic analysis basic information, space type characteristic analysis basic information, price type characteristic analysis basic information, storage type characteristic analysis basic information and quality type characteristic analysis basic information in each commodity to form commodity sales characteristic data; according to the historical sales data of the single product, performing characteristic analysis aiming at time-class sales characteristics affecting commodity sales, and forming time-class characteristic analysis basic information, wherein the method comprises the following steps: acquiring n characteristic sales cycles of different commodities in target storeThe sales data in the price class sales feature is filtered to eliminate the influence of the price class sales feature on sales volume, and a feature sales period data set +.>The method comprises the steps of carrying out a first treatment on the surface of the Wherein, each characteristic sales period data +.>Corresponding time period +.>The sales amount is->，/>M is->Sequential numbering of the divided time periods in the time dimension order; determining the characteristic sales cycle of the commodity according to the characteristic sales cycle data set A>Average sales per time period +.>The method comprises the steps of carrying out a first treatment on the surface of the The characteristic sales period is +.>All of said average sales +.>Forming the time-class characteristic analysis basic information of different commodities; according to the historical sales data of the single product, performing characteristic analysis aiming at space sales characteristics affecting commodity sales, and forming space characteristic analysis basic information, wherein the method comprises the following steps:

Setting a space analysis area, taking the target store as a position base point to sell with the target storeThe number of correlation stores of the same target commodity is different as an extraction target, and the characteristic sales periods with different correlation stores are extracted from n characteristic sales periods, and are determined as space analysis periods; grouping the space analysis periods according to the sequence from the near to the far of the time dimension to form space analysis period groups, ensuring that the number of the space analysis periods contained in each group is 3, and removing the rest space analysis periods if the number of the rest space analysis periods is less than 3; for each space analysis period, establishing a two-dimensional position coordinate system by taking the target store as an origin, and determining the distance between the correlation store and the target storeU is the number of the corresponding correlation store and the relative space rate is determined>The method comprises the steps of carrying out a first treatment on the surface of the Determining a sales amount +/for each of the correlation stores in the corresponding spatial analysis period>And determining the average sales amount +.>The method comprises the steps of carrying out a first treatment on the surface of the Determining a target sales amount of the target commodity of the target store in the corresponding space analysis period >The method comprises the steps of carrying out a first treatment on the surface of the According to the relative spatial rate->Said average sales amount->The target sales amount->And the number u of the correlation stores, determining a spatial constraint parameter corresponding to the spatial analysis period>Wherein->The method comprises the steps of carrying out a first treatment on the surface of the Three space constraint parameters in each space analysis period group are obtained, an objective function of h is established, and the relative space rate +_is used for>And the number u of the correlation stores is a binary function of the variables; homogenizing according to the binary function of each space analysis period group to form space constraint function +.>The method comprises the steps of carrying out a first treatment on the surface of the And combining the space constraint functions of all commodities in the target store to form the space class characteristic analysis basic information.

2. The method for intelligent commodity matching and matching according to claim 1, wherein said performing characteristic analysis for price class sales characteristics affecting commodity sales according to said historical sales data of single commodity to form price class characteristic analysis basic information comprises:

determining a price analysis area in which uniform discount values for target goods of different stores in n characteristic sales cycles are obtainedSaid unified discount value- >Corresponding unified discount period->Uniform discount sales->V is the number corresponding to the store selling the target commodity in the characteristic sales period with the number x, x represents the +_ with the unified discount period extracted in n characteristic sales periods>A number of the characteristic sales period of (2);

analyzing basic information according to the time class characteristics, and determining the uniform discount periodAdjacent and at the unified discounts period->Average sales in the preceding time period +.>And +/with the unified discount period>Adjacent and at the unified discounts period->Average sales>；

Determining an average uniform discount rate of change according to：

，/>Wherein->；

Acquiring in-store discount values for the target commodity in n characteristic sales cycles for different storesSaid in-store discount value +.>Corresponding in-store discount period->In-store discount sales->I represents the number of the store selling the target commodity, and y represents the number of the characteristic sales cycle for in-store discounts by the store with the number i;

analyzing basic information according to the time-class characteristics to determine discount periods of different shops in the shopsCorresponding to said characteristic sales period +.>Adjacent and in said in-store discount period- >Average sales in the preceding time period +.>And +.>Corresponding to said characteristic sales period +.>Adjacent and in said in-store discount period->Average sales>；

Determining average in-store discount rate of change for different stores according to：

，/>，/>；

Combining the average uniform discounted rate of changeAnd all of said average in-store discount rate of change +.>And forming the price class characteristic analysis basic information.

3. The method for intelligent commodity matching and matching according to claim 2, wherein said performing characteristic analysis for storage class sales characteristics affecting commodity sales according to said commodity historical sales data to form storage class characteristic analysis basic information comprises:

acquiring the residual stock quantity of target commodities in different shops in n characteristic sales periodsAnd corresponding sales ∈ ->Z represents a store number for selling the target commodity, and k is a sequence number in a time dimension of the characteristic sales cycles for selling the target commodity extracted in n characteristic sales cycles;

based on the remaining stock quantityAnd said sales +.>Determining the pin memory ratio->Wherein:；

according to the pin-to-memory ratio Determining a pin ratio difference between adjacent ones of said characteristic sales periods in a time dimension +.>，2≤/>≤k；

According to the pin ratio differenceEstablishing a time-dimension-based function of the rate of change of the sales of the target commodity in different stores +.>；

Combining all of the pin deposit rate of change functionsAnd forming the storage class characteristic analysis basic information.

4. A method for intelligent commodity matching and matching according to claim 3, wherein said performing characteristic analysis for quality class sales characteristics affecting commodity sales according to said historical sales data of said single commodity to form quality class characteristic analysis basic information comprises:

analyzing basic information according to the time-class characteristics to obtain sales amounts of target commodities in each of n characteristic sales periodsAnd sales amount of the same type of goods as the target goods +.>E represents the number of the same type of goods as the target goods, wherein ∈10>，；

In each characteristic sales period, obtaining sales differences of the similar commodities and the target commodityAnd according to the sales difference, establishing a sales difference change rate function of the target commodity relative to the similar commodity based on time dimension +. >；

Combining all of said sales difference change rate functions of said target commodity relative to said like commodityAnd forming the quality class characteristic analysis basic information.

5. The intelligent commodity replenishment method according to claim 4, wherein said determining the amount of orders for different commodities for each store based on said commodity sales characteristic data, forming store order data, comprises:

determining the goods s sold in the store and determining the order amount of each of the goods in the store according to the following analysis and judgment stepsS is the number of all different kinds of commodities:

setting a sales difference rate thresholdAccording to the sales difference rate function +.>Determining the sales difference change rate of the commodity s relative to the similar commodity in the next characteristic sales period>And makes the following decisions:

if it isThe store no longer supplies items s during the next said characteristic sales cycle;

if it isThe store is next saidContinuing to supply the goods s in the characteristic sales cycle, and determining the order quantity of the goods s according to the following manner>：

Determining the total average sales of the commodity in the characteristic sales period according to the time-class characteristic analysis basic information Wherein: />；

Acquiring the number u of stores and the relative space rate of the same goods s sold by the stores in the next characteristic sale periodAccording to the space constraint function->Determining the space restriction rate of commodity s>；

Acquiring a preset unified discount period of the store in the next characteristic sales periodAnd preset unified discount value->And according to said average unified discount rate of change of commodity s +.>Determining that commodity s is +.>Sales amount of->Wherein: />；

Acquiring a preset in-store discount period of a store in a next characteristic sales periodAnd preset in-store discount value->And +/in accordance with the preset in-store discount period of the item s>Said preset in-store discount value->Said preset unified discount value +.>Determining that merchandise s is +/in the preset in-store discount period>Sales amount of->Wherein:

at this time i is +.>Corresponding store numbers;

setting a pin ratio thresholdAccording to the pin memory change rate function +.>Determining the stock-out ratio of the products s in the next said characteristic sales cycle in the store +.>And makes the following decisions:

if it isThen ∈the stock ratio threshold ∈>As the stock/sales ratio of the commodity s in the next characteristic sales period, and determining the residual storage amount +. >Wherein:

；

if it isAccording to->Determining the remaining memory of the item s in the next said characteristic sales period>Wherein: />；

According to the total average sales of the goods sSaid spatial restriction rate->Said preset unified discounts period +.>Sales amount of->Said preset in-store discount period->Sales amount of->Said residual memory +.>Determining said order amount of goods s in store +.>Wherein: />；

According to the order of different products s in storeAnd determining the store order quantity data.

6. The intelligent commodity replenishment method according to claim 5, wherein said determining actual order information based on said store order quantity data and commodity replenishment for different stores based on said actual order information, forming store replenishment data, comprises:

acquiring the store nextA sales task amount characterizing a sales cycle and based on the order amount of the product sDetermining additional turnover of store +.>；

Determining the store first shipment data based on the order volume and the additional turnover volume for different items within the store, wherein:

acquiring sales task amount of store in next characteristic sales period and ordering goods according to the goods s Determining additional turnover of store +.>Comprising:

acquiring the sales task amount of the store in the next characteristic sales period, and determining the residual memory capacity of the store by combining the ordering amount of different commodities in the store;

determining the total ordering amount of the same commodity s in the next characteristic sales period according to the ordering amount of the same commodity s in different stores, sorting the commodity s from large to small according to the total ordering amount, extracting the total ordering amount of the previous p commodities, and determining the total ordering amount ratio of the p commodities;

and acquiring the storage amounts of all shops, and respectively giving the storage amounts to p commodities according to the total order amount ratio.

7. The method of intelligent merchandise allocation and replenishment tuning according to claim 6, wherein the acquiring real-time merchandise volume information of a store and performing replenishment tuning analysis in combination with the first merchandise allocation data of the store to form replenishment tuning analysis data comprises:

setting a stock early warning lower limit valueAnd stock early warning upper limit +.>And->；

Acquiring real-time stock of all commodities in a store, and when the real-time stock of the commodities is not more than the stock early warning lower limit valueWhen the stores are used as position points, the stock in other stores is determined to exceed the stock early warning upper limit value +_in sequence from near to far according to the relative distance >Extracting +.about.upper limit of stock early warning from different stores in order of relative distance from near to far>Until the accumulated and extracted stock guarantees that the stock reaches the stock early warning upper limit value +_after store supplementation>Until that point.