Patent Application
20240037496
This U.S. patent application claims priority under 35 U.S.C. § 119 to: Indian Patent Application No. 202221042736, filed on Jul. 26, 2022. The entire contents of the aforementioned application are incorporated herein by reference.
The embodiments herein generally relate to the field of retail store management and, more particularly, to a method and system for space planning providing techniques to create floor plans and planograms across retail stores of a retailer.
In retail, space planning is carried out at corporate level and there are challenges in getting holistic view of space optimization at corporate level of retailer. Retailers such as Walmart or the like have enormous number of stores and macro space optimization is carried out at individual store level. Space optimization tools have provision to run space optimization for all the stores simultaneously and they recommend new space for each category for each store. A category may be defined as group of products that are having attribute values that are alternative to each other. Example for category are women's tops, women's suits, men's shirts, and men's suits, mobile, dairy, etc., A department is group of categories that meet related needs. Example for department are beverages, cereals, snacks, etc., Each store may behave differently with respect to space allocation. Based on corporate need, macro space optimization is carried out for the objective of expand space, reduce space and constant space separately. Due to these procedures, the number of space recommendations analyzed at corporate level increases extremely high making it difficult to bring key inferences out of these recommendations. Analyst at Head Quarters (HQ) of the retailer faces challenges in getting holistic view of space optimization.
Further, based on corporate strategy such as expansion or reduction of total store space, planograms are created. Major retailers create planogram at individual store level mostly or in some instances at store cluster level to save effort, time, and cost. These clusters are formed based on store size, store format, demographic, competition, etc., and the approach faces challenges in defining the trade area to derive demographic variable, competition, and others. It is challenging to define trade area. In an instance deciding competition intensity based on trade area is a challenging and subjective one. For example, for a general merchant retailer, the trade area for food related categories of a store could be 5 miles whereas for a laptop category it could be 30 miles since the customers might be willing to travel different distances depending upon the category. Due to these factors, clustering based on sales drivers such as demo, competition, weather, etc. has challenges and provides inaccurate end results.
Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems.
For example, in one embodiment, a method for space planning is provided. The method includes acquiring store-data over a predefined time span from a plurality of stores and obtaining sales, space, and demographic information from the store-data on a plurality of categories in the plurality of stores. Further, the method includes processing the sales, space, and demographic information, via a space optimization tool implemented by the one or more hardware processors, to perform space optimization in accordance with a set of predefined optimization rules comprising category level optimization rules and in aisle optimization rules, to generate: (a) a plurality of final space allocations for each of the plurality of categories of each of the plurality of stores; (b) a plurality of delta space allocations for each of the plurality of categories of each of the plurality of stores, wherein a delta space is defined as a difference between an initial space occupied by a category amongst the plurality of categories and a final space suggested by the space optimization tool for the category; and (c) details on a space allocation order for the category for each incremental foot during each iteration of space optimization, by-the space optimization tool, captured in a plurality of log files for a plurality of stores.
Further, the method includes creating a plurality of vectors comprising: (a) a final space vector capturing a final space allocation corresponding to each of the plurality of categories of a store among the plurality of stores, wherein the final space allocation corresponding to each of the plurality of stores is processed in a single row vector format as the final space vector; (b) a delta space vector capturing a delta space allocation for each of the plurality of categories for the store, wherein the delta space allocation corresponding to each of the plurality of stores is processed in a single row vector format as the delta space vector; and (c) an allocation order vector capturing order of priority in space allocation from the space allocation order across the plurality of categories for the store, wherein a log file for a store from amongst the plurality of log files is processed into a plurality of row vectors with same length noted as ‘basic allocation order vectors’, wherein each row vector represents an iteration of optimization-, and number of row vectors is equal to number of iterations and length of each row vector is equal to number of the plurality of categories of corresponding store, each cell of the row represents a category among the plurality of categories, and value of each cell of the row represents a fixed incremental space allocation for the category for an incremental foot during the iteration, processing of basic allocation order vectors is carried out by calculating cumulative totals for each iteration and termed as ‘cumulative total allocation order vectors’ and further processing is carried out by adding basic allocation order vectors or cumulative total allocation order vectors one by one to form a single row vector representing the allocation order vector in two formats namely basic allocation order vector and cumulative allocation order vector.
Furthermore, the method includes generating a plurality of space matrices wherein each column represents the store and number of columns are equivalent to number of the plurality of stores, and wherein each row represents a category and number of rows equivalent to a number of the plurality of categories, the plurality of space matrices comprising: (a) a final space matrix, generated from the final space vector corresponding to each of the plurality of stores, by converting the format of the final space vector from a single row vector to a single column vector, and considering the plurality of the stores and arranging a plurality of final space vectors of the plurality of stores into a matrix format, wherein value of each element of the column vector represents final space allocated to each of the plurality of categories; (b) delta space matrix, generated from the delta space vector corresponding to each of the plurality of stores, by converting format of the delta space vector from a single row vector to a single column vector, and considering the plurality of the stores and arranging a plurality of delta space vectors of the plurality of stores into a matrix format; and (c) a space allocation order matrix, generated from the allocation order vector corresponding to each of the plurality of stores, by converting the allocation order vector format to a column vector format and considering the plurality of stores and arranging a plurality of allocation order vectors of the plurality of stores into a matrix format.
Further, the method includes processing via a pattern extraction tool implemented by the one or more hardware processors the final space matrix, the delta space matrix and the space allocation order matrix by using a plurality of formats of measurements of final space, delta space, and the space allocation order and applying standardization, a covariance matrix creation, and principal component analysis (PCA) to identify a set of patterns for each of the final space matrix, the delta space matrix and the space allocation order matrix based on eigen vectors and eigen values generated during the (PCA). Further, the method includes generating a store level mismatch score by comparing existing floor plans with the set of patterns derived from each of the final space matrix, the delta space matrix and the space allocation order matrix, and processing the set of patterns for each of the final space matrix, the delta space matrix and the space allocation order matrix under the plurality of formats of measurements of the final space, the delta space, and the space allocation order to determine quality of the set of patterns from variance contribution of the set of patterns of PCA.
Furthermore, the method includes performing three level iterations comprising: (a) a first level of iterations to select top set of patterns from the set of patterns formed from each of (i) the final space matrix (ii) the delta space matrix, and (iii) the allocation order, under different formats of final space, delta space, and allocation order and selecting a top set of pattern for a suitable format among one of the final space, the delta space, and the space allocation order, wherein the quality of set of patterns is maximum for the top set of pattern; (b) a second level of iterations to apply to set of patterns received from the first level iterations in which outcome of each iteration within second level of iterations is used to locate a mismatch store using the mismatch score and recalculating any one of the (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation order for only located mismatching store using identified rules based on outcome of previous iteration within second level of iterations, and generating the set of patterns by adding one of the recalculated (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation order, wherein iteration continues until no store is deviating from the pattern in terms of aisle fitment; and (c) a third level of iterations applied to set of patterns received from second level iterations in which outcome of each iteration within the third level of iterations is used to locate the store deviating from the set of patterns using multivariate distance, and recalculating any one of the (i) the final space matrix (ii) the delta space matrix, and (iii) the allocation order for only those deviating store using identified rules based on outcome of previous iteration within third level of iterations, and generating set of patterns by adding one of the recalculated (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation, wherein iteration continues until no store is deviating from the pattern.
Further, the method includes generating a set of floorplans and planograms in accordance with the set of patterns received from third level iterations, wherein the floorplans and planograms are recommended to each of the plurality of stores in accordance with set of patterns received from third level iterations.
In another aspect, a system for space planning is provided. The system comprises a memory storing instructions; one or more Input/Output (I/O) interfaces; and one or more hardware processors coupled to the memory via the one or more I/O interfaces, wherein the one or more hardware processors are configured by the instructions to acquire store-data over a predefined time span from a plurality of stores and obtaining sales, space, and demographic information from the store-data on a plurality of categories in the plurality of stores. Further, the system processes the sales, space, and demographic information, via a space optimization tool implemented by the one or more hardware processors, to perform space optimization in accordance with a set of predefined optimization rules comprising category level optimization rules and in aisle optimization rules, to generate: (a) a plurality of final space allocations for each of the plurality of categories of each of the plurality of stores; (b) a plurality of delta space allocations for each of the plurality of categories of each of the plurality of stores, wherein a delta space is defined as a difference between an initial space occupied by a category amongst the plurality of categories and a final space suggested by the space optimization tool for the category; and (c) details on a space allocation order for the category for each incremental foot during each iteration of space optimization, by-the space optimization tool, captured in a plurality of log files for a plurality of stores.
Further, system creates a plurality of vectors comprising: (a) a final space vector capturing a final space allocation corresponding to each of the plurality of categories of a store among the plurality of stores, wherein the final space allocation corresponding to each of the plurality of stores is processed in a single row vector format as the final space vector; (b) a delta space vector capturing a delta space allocation for each of the plurality of categories for the store, wherein the delta space allocation corresponding to each of the plurality of stores is processed in a single row vector format as the delta space vector; and (c) an allocation order vector capturing order of priority in space allocation from the space allocation order across the plurality of categories for the store, wherein a log file for a store from amongst the plurality of log files is processed into a plurality of row vectors with same length noted as ‘basic allocation order vectors’, wherein each row vector represents an iteration of optimization-, and number of row vectors is equal to number of iterations and length of each row vector is equal to number of the plurality of categories of corresponding store, each cell of the row represents a category among the plurality of categories, and value of each cell of the row represents a fixed incremental space allocation for the category for an incremental foot during the iteration, processing of basic allocation order vectors is carried out by calculating cumulative totals for each iteration and termed as ‘cumulative total allocation order vectors’ and further processing is carried out by adding basic allocation order vectors or cumulative total allocation order vectors one by one to form a single row vector representing the allocation order vector in two formats namely basic allocation order vector and cumulative allocation order vector.
Furthermore, the system generates a plurality of space matrices wherein each column represents the store and number of columns are equivalent to number of the plurality of stores, and wherein each row represents a category and number of rows equivalent to a number of the plurality of categories, the plurality of space matrices comprising: (a) a final space matrix, generated from the final space vector corresponding to each of the plurality of stores, by converting the format of the final space vector from a single row vector to a single column vector, and considering the plurality of the stores and arranging a plurality of final space vectors of the plurality of stores into a matrix format, wherein value of each element of the column vector represents final space allocated to each of the plurality of categories; (b) delta space matrix, generated from the delta space vector corresponding to each of the plurality of stores, by converting format of the delta space vector from a single row vector to a single column vector, and considering the plurality of the stores and arranging a plurality of delta space vectors of the plurality of stores into a matrix format; and (c) a space allocation order matrix, generated from the allocation order vector corresponding to each of the plurality of stores, by converting the allocation order vector format to a column vector format and considering the plurality of stores and arranging a plurality of allocation order vectors of the plurality of stores into a matrix format.
Further, the system processes via a pattern extraction tool implemented by the one or more hardware processors the final space matrix, the delta space matrix and the space allocation order matrix by using a plurality of formats of measurements of final space, delta space, and the space allocation order and applying standardization, a covariance matrix creation, and principal component analysis (PCA) to identify a set of patterns for each of the final space matrix, the delta space matrix and the space allocation order matrix based on eigen vectors and eigen values generated during the (PCA). Further, the system generates a store level mismatch score by comparing existing floor plans with the set of patterns derived from each of the final space matrix, the delta space matrix and the space allocation order matrix, and processing the set of patterns for each of the final space matrix, the delta space matrix and the space allocation order matrix under the plurality of formats of measurements of the final space, the delta space, and the space allocation order to determine quality of the set of patterns from variance contribution of the set of patterns of PCA.
Furthermore, the system performs three level iterations comprising: (a) a first level of iterations to select top set of patterns from the set of patterns formed from each of (i) the final space matrix (ii) the delta space matrix, and (iii) the allocation order, under different formats of final space, delta space, and allocation order and selecting a top set of pattern for a suitable format among one of the final space, the delta space, and the space allocation order, wherein the quality of set of patterns is maximum for the top set of pattern; (b) a second level of iterations to apply to set of patterns received from the first level iterations in which outcome of each iteration within second level of iterations is used to locate a mismatch store using the mismatch score and recalculating any one of the (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation order for only located mismatching store using identified rules based on outcome of previous iteration within second level of iterations, and generating the set of patterns by adding one of the recalculated (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation order, wherein iteration continues until no store is deviating from the pattern in terms of aisle fitment; and (c) a third level of iterations applied to set of patterns received from second level iterations in which outcome of each iteration within the third level of iterations is used to locate the store deviating from the set of patterns using multivariate distance, and recalculating any one of the (i) the final space matrix (ii) the delta space matrix, and (iii) the allocation order for only those deviating store using identified rules based on outcome of previous iteration within third level of iterations, and generating set of patterns by adding one of the recalculated (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation, wherein iteration continues until no store is deviating from the pattern.
Further, the system generates a set of floorplans and planograms in accordance with the set of patterns received from third level iterations, wherein the floorplans and planograms are recommended to each of the plurality of stores in accordance with set of patterns received from third level iterations.
In yet another aspect, there are provided one or more non-transitory machine-readable information storage mediums comprising one or more instructions, which when executed by one or more hardware processors causes a method for space planning. The method includes acquiring store-data over a predefined time span from a plurality of stores and obtaining sales, space, and demographic information from the store-data on a plurality of categories in the plurality of stores. Further, the method includes processing the sales, space, and demographic information, via a space optimization tool implemented by the one or more hardware processors, to perform space optimization in accordance with a set of predefined optimization rules comprising category level optimization rules and in aisle optimization rules, to generate: (a) a plurality of final space allocations for each of the plurality of categories of each of the plurality of stores; (b) a plurality of delta space allocations for each of the plurality of categories of each of the plurality of stores, wherein a delta space is defined as a difference between an initial space occupied by a category amongst the plurality of categories and a final space suggested by the space optimization tool for the category; and (c) details on a space allocation order for the category for each incremental foot during each iteration of space optimization, by-the space optimization tool, captured in a plurality of log files for a plurality of stores.
Further, the method includes creating a plurality of vectors comprising: (a) a final space vector capturing a final space allocation corresponding to each of the plurality of categories of a store among the plurality of stores, wherein the final space allocation corresponding to each of the plurality of stores is processed in a single row vector format as the final space vector; (b) a delta space vector capturing a delta space allocation for each of the plurality of categories for the store, wherein the delta space allocation corresponding to each of the plurality of stores is processed in a single row vector format as the delta space vector; and (c) an allocation order vector capturing order of priority in space allocation from the space allocation order across the plurality of categories for the store, wherein a log file for a store from amongst the plurality of log files is processed into a plurality of row vectors with same length noted as ‘basic allocation order vectors’, wherein each row vector represents an iteration of optimization-, and number of row vectors is equal to number of iterations and length of each row vector is equal to number of the plurality of categories of corresponding store, each cell of the row represents a category among the plurality of categories, and value of each cell of the row represents a fixed incremental space allocation for the category for an incremental foot during the iteration, processing of basic allocation order vectors is carried out by calculating cumulative totals for each iteration and termed as ‘cumulative total allocation order vectors’ and further processing is carried out by adding basic allocation order vectors or cumulative total allocation order vectors one by one to form a single row vector representing the allocation order vector in two formats namely basic allocation order vector and cumulative allocation order vector.
Furthermore, the method includes generating a plurality of space matrices wherein each column represents the store and number of columns are equivalent to number of the plurality of stores, and wherein each row represents a category and number of rows equivalent to a number of the plurality of categories, the plurality of space matrices comprising: (a) a final space matrix, generated from the final space vector corresponding to each of the plurality of stores, by converting the format of the final space vector from a single row vector to a single column vector, and considering the plurality of the stores and arranging a plurality of final space vectors of the plurality of stores into a matrix format, wherein value of each element of the column vector represents final space allocated to each of the plurality of categories; (b) delta space matrix, generated from the delta space vector corresponding to each of the plurality of stores, by converting format of the delta space vector from a single row vector to a single column vector, and considering the plurality of the stores and arranging a plurality of delta space vectors of the plurality of stores into a matrix format; and (c) a space allocation order matrix, generated from the allocation order vector corresponding to each of the plurality of stores, by converting the allocation order vector format to a column vector format and considering the plurality of stores and arranging a plurality of allocation order vectors of the plurality of stores into a matrix format.
Further, the method includes processing via a pattern extraction tool implemented by the one or more hardware processors the final space matrix, the delta space matrix and the space allocation order matrix by using a plurality of formats of measurements of final space, delta space, and the space allocation order and applying standardization, a covariance matrix creation, and principal component analysis (PCA) to identify a set of patterns for each of the final space matrix, the delta space matrix and the space allocation order matrix based on eigen vectors and eigen values generated during the (PCA). Further, the method includes generating a store level mismatch score by comparing existing floor plans with the set of patterns derived from each of the final space matrix, the delta space matrix and the space allocation order matrix, and processing the set of patterns for each of the final space matrix, the delta space matrix and the space allocation order matrix under the plurality of formats of measurements of the final space, the delta space, and the space allocation order to determine quality of the set of patterns from variance contribution of the set of patterns of PCA.
Furthermore, the method includes performing three level iterations comprising: (a) a first level of iterations to select top set of patterns from the set of patterns formed from each of (i) the final space matrix (ii) the delta space matrix, and (iii) the allocation order, under different formats of final space, delta space, and allocation order and selecting a top set of pattern for a suitable format among one of the final space, the delta space, and the space allocation order, wherein the quality of set of patterns is maximum for the top set of pattern; (b) a second level of iterations to apply to set of patterns received from the first level iterations in which outcome of each iteration within second level of iterations is used to locate a mismatch store using the mismatch score and recalculating any one of the (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation order for only located mismatching store using identified rules based on outcome of previous iteration within second level of iterations, and generating the set of patterns by adding one of the recalculated (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation order, wherein iteration continues until no store is deviating from the pattern in terms of aisle fitment; and (c) a third level of iterations applied to set of patterns received from second level iterations in which outcome of each iteration within the third level of iterations is used to locate the store deviating from the set of patterns using multivariate distance, and recalculating any one of the (i) the final space matrix (ii) the delta space matrix, and (iii) the allocation order for only those deviating store using identified rules based on outcome of previous iteration within third level of iterations, and generating set of patterns by adding one of the recalculated (i) the final space matrix (ii) the delta space matrix, and (iii) the space allocation, wherein iteration continues until no store is deviating from the pattern.
Further, the method includes generating a set of floorplans and planograms in accordance with the set of patterns received from third level iterations, wherein the floorplans and planograms are recommended to each of the plurality of stores in accordance with set of patterns received from third level iterations. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles:
7B depict aisle mismatch score for a store based on the extent of how the categories are underfilled or overfilled for the store when they are fitted in aisle with optimized space recommendation for the store, in accordance with some embodiments of the present disclosure.
It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems and devices embodying the principles of the present subject matter. Similarly, it will be appreciated that any flow charts, flow diagrams, and the like represent various processes which may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.
Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments.
In retail, macro space optimization is carried out at individual stores to allocate optimum space for each category. As each retailer has many stores and macro space optimization is experimented with different objectives such as expand space, reduce space and constant space of a store individually, the number of space recommendations analyzed at corporate level increases extremely high making it difficult to bring key inferences out of these recommendations and creating challenges in implementation of results such as creation of planograms and floor plans.
Embodiments of the present disclosure provide a method and system for space planning providing techniques to create floor plans and planograms. The method identifies underlying patterns that reside in space recommendations across stores and creation of optimal number of floor plans and planograms in accordance with the identified set of patterns unlike large number of floor plans or planograms generated by state of the art space planning systems.
Referring now to the drawings, and more particularly to
The system 100, via a pattern extraction tool, extracts patterns from generated space recommendations by suitable data formats and using data mining techniques. such as Principal Component Analysis (PCA). Factors that are responsible for pattern formation are identified by mapping those patterns with available sales drivers and applying suitable multivariate models such as canonical discriminant analysis. Thus, with the system 100, analyst at corporate office of the retailer is able to get top patterns with their magnitude and potential causes to understand the process better and to make strategies effectively.
Further, the system 100 also creates planogram for each rational group of stores based on pattern detected.
As patterns are derived from hidden relationship across stores, it has the better insights. The patterns are formed based on space allocation logic which varies across stores, and which is based on location of the store and other causative factors such as competition, promotion, etc., the patterns formed from space allocation are useful to be used for practical applications such as remodeling of stores or resetting of stores. Remodeling of stores refers to major changes which is done for once in many years and it is usually carried out by changing floor plans and changing racks, etc. As floor planning execution cannot be altered frequently and involves huge cost impact in terms of labor, procurement of infrastructure, fixtures, product inventory and other store remodel aspects, it is given utmost importance and considered as one of the primary functions of retail merchandise planning. Floor planning is the process of arranging the categories and department within the retail store floor based on space allocated to the respective department and categories. Resetting of stores refers to minor changes which is done, once in 3 months and it is usually carried out by changing planograms. Floor planning happens at a macro level post finalizing which the micro level planning of planogram, fixture, and item into this floor space happens. Floor planning happens at the corporate level for each retail store guided by space planner, analyst, category managers and store managers.
Referring to the components of system 100, in an embodiment, the processor(s) 104, can be one or more hardware processors 104. In an embodiment, the one or more hardware processors 104 can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the one or more hardware processors 104 are configured to fetch and execute computer-readable instructions stored in the memory 102. In an embodiment, the system 100 can be implemented in a variety of computing systems including laptop computers, notebooks, hand-held devices such as mobile phones, workstations, mainframe computers, servers, and the like.
The I/O interface(s) 106 can include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface to display the generated target images and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular and the like. In an embodiment, the I/O interface (s) 106 can include one or more ports for connecting to a number of external devices or to another server or devices.
The memory 102 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.
In an embodiment, the memory 102 includes a plurality of modules 110 such as a pattern extraction tool, a space optimization tool (not shown). The plurality of modules 110 further include programs or coded instructions that supplement applications or functions performed by the system 100 for executing different steps involved in the process of space planning providing optimal space recommendations for categories across retail stores of the retailer, being performed by the system 100. The plurality of modules 110, amongst other things, can include routines, programs, objects, components, and data structures, which performs particular tasks or implement particular abstract data types. The plurality of modules 110 may also be used as, signal processor(s), node machine(s), logic circuitries, and/or any other device or component that manipulates signals based on operational instructions. Further, the plurality of modules 110 can be used by hardware, by computer-readable instructions executed by the one or more hardware processors 104, or by a combination thereof. The plurality of modules 110 can include various sub-modules (not shown). Further, the memory 102 may comprise information pertaining to input(s)/output(s) of each step performed by the processor(s) 104 of the system 100 and methods of the present disclosure.
Further, the memory 102 includes a database 108. The database (or repository) 108 may include a plurality of abstracted piece of code for refinement and data that is processed, received, or generated as a result of the execution of the plurality of modules in the module(s) 110. It can store space recommendations, identified patterns from the space recommendation, planograms, floor plans and the like. Although the data base 108 is shown internal to the system 100, it will be noted that, in alternate embodiments, the database 108 can also be implemented external to the system 100, and communicatively coupled to the system 100. The data contained within such external database may be periodically updated. For example, new data may be added into the database (not shown in
In an embodiment, the system 100 comprises one or more data storage devices or the memory 102 operatively coupled to the processor(s) 104 and is configured to store instructions for execution of steps of the method 300 by the processor(s) or one or more hardware processors 104. The steps of the method 300 of the present disclosure will now be explained with reference to the components or blocks of the system 100 as depicted in
Referring to the steps of the method 300, at step 302 of the method 300, the one or more hardware processors 104 acquire store-data over a predefined time span from a plurality of stores. At step 304 of the method 300, the one or more hardware processors 104 obtain sales, space, and demographic information from the store-data on a plurality of categories in the plurality of stores. Information associated with space optimization namely sales, space and demographic information for a predefined period or time span is received from various sources. POS (Point Of Sales) systems capture category level sales information. Historical space allocated to categories of a store are received from historical planograms. Demographic information about the stores are received from third party vendors and stored in respective systems.
At step 306 of the method 300, the one or more hardware processors 104 processes the sales, space, and demographic information, via the space optimization tool performing space optimization in accordance with a set of predefined optimization rules comprising category level optimization rules and in aisle optimization rules, to generate the following:
The optimization rules refer to guidelines that need to be followed by a space optimization algorithm to address retail strategies and to avoid implementation issues and the optimization results need to adhere to these guidelines. The rules may be predefined or dynamic which is explained in detail in latter part. Many types of optimization rules are followed during optimization.
Few examples are mentioned below:
Aisle level complexities may arise in diverse ways. In an instance, multiple departments may be placed within same aisle, such as right side of an aisle may occupy department1 and first of left side of the aisle may be occupied by department2 and rest of left side may be occupied by department 3. Hence more aisle category level space limits need to be applied for each of the department. In another instance, Same departments are placed across multiple aisles fully or partially (E.g., Dry Grocery departments are placed in aisle1, aisle2, half of aisle3 and making the size as 2.5 aisle). At step 308 of the method 300, the one or more hardware processors 104 create a plurality of vectors comprising:
The space optimization tool provides space recommendations for a store. It has information on the new space that is recommended to each category of the store. The recommendations are based on the past historical performance of the categories which varies based on the demographic information associated with the store and historical space allocated to the store and comparative performance of the categories.
The final space—The space optimization tool has provision to run space optimization for all the stores simultaneously by applying predefined optimization rules and they recommend new space for each category for each store. For example, the new space for a category may be in different format such as final space in actual units, delta space in actual units, delta space in percentage, delta space in percentage and many more format which is within the scope of present disclosure. Space for a category is measured in different formats such as linear feet, square feet, weighted visible space, cognitive visible space and many more format which is within the scope of present disclosure. It is assumed that sales of a category is related to space allocated to the category. In other words, sales of a category is related to amount of space that is visible for customers during purchase in a store and the space may be termed as visible space. Different format of category space is discussed as below:
The final space allocated to each categories is processed in a single row vector format as displayed in the
The delta space—The delta space is defined as the space difference between initial space occupied by the category and final space suggested by space optimization tool. Delta space may be defined in different form such as original units i.e., total additional feet, percentage of additional space as compared with initial space occupied by the category or percentage of additional space as compared with available space for the category. Here available space is decided by the optimization rules such as minimum space and maximum space for the category in which minimum space is 80% of current space and maximum space is 120% of current space. If the current space of a category is 500 feet and the minimum space is 400 feet and the maximum space is 600. So, the available space is 100 feet in both direction from 500 feet. If the recommended space is 570 feet, then 70% of available space is occupied by the category. Total additional feet allocated to each categories is processed in a single row vector format as displayed in the
The allocation order—Log files of each store from space optimization tool are fetched. Log files of each optimization have the details of each incremental space allocation that occur during each iteration of optimization which is carried out using same business rules across stores. A log file for a store is processed into row vectors with same length noted as ‘basic allocation order vectors’ as displayed in
Basic allocation order vector is processed by calculating cumulative totals for each iteration and termed as ‘cumulative total allocation order vectors’ as displayed in
Further processing is carried out by adding basic allocation order vectors or cumulative total allocation order vectors one by one to form a single row vector representing the allocation order vector in two formats namely basic allocation order vector and cumulative allocation order vector. Cumulative allocation order vector in single row is displayed in
At step 310 of the method 300, the one or more hardware processors 104 generate a plurality of space matrices, wherein each column represents the store and number of columns are equivalent to number of the plurality of stores, and wherein each row represents a category and number of rows equivalent to a number of the plurality of categories. The plurality of space matrices comprising:
As depicted in
Once the three types of matrices are generated, at step 312 of the method 300, the one or more hardware processors 104, via the pattern extraction tool, process the final space matrix, the delta space matrix and the space allocation order matrix by using a plurality of formats of measurements of final space, delta space, and the space allocation order and applying standardization, a covariance matrix creation, and principal component analysis (PCA) to identify a set of patterns for each of the final space matrix, the delta space matrix and the space allocation order matrix based on eigenvectors and eigenvalues generated during the (PCA).
Each matrix undergoes into standardization, creation of covariance matrix, and principal component analysis. In an embodiment, standardization is carried out by subtracting the individual store value with mean store values i.e., each store vector is subtracted from the vector of mean value in which the vector of mean value is calculated by considering all the stores for every cell. In an embodiment, the standardized matrix includes ‘NI’ columns and ‘n’ rows. As can be understood, standardization enables bringing the numeric values of each store to a common/standard scale for further processing. Standardized matrix is processed to form covariance matrix. As mentioned, the Principal component analysis (PCA) is applied on the covariance matrix to generate the set of eigenvectors with corresponding eigenvalues and factor loadings in which each eigenvector which are greater than 1 are considered as pattern. The PCA when applied on the covariance matrix generates the set of eigenvectors with corresponding eigenvalues and factor loadings. Eigenvalues decrease exponentially and those eigenvectors which are greater than 1 are considered as patterns and they are noted as set of patterns. Quality of the set of patterns is determined from variance contribution of the set of patterns of PCA.
Quality of the set of patterns=sum of variance contribution of the set of patterns or eigenvectors which are greater than 1.
Each store is tagged into an underlying pattern based on factor loadings. Each underlying pattern is arrived based on common characteristics that are available across many stores. Each underlying pattern is arrived based on common characteristics across stores in terms of space optimization mechanism such as priority of space capturing by the categories, interaction between categories due to different space elasticity while capturing space by categories, simultaneous gaining of spaces by similar productivity categories to attain global optimum, and other backend mechanism that are happening during space optimization.
Patterns are generated based on (i) final space allocations (ii) delta space allocations and (iii) space allocation logic and based on different format of measurements. Each combination produces a set of patterns. For example, final space allocations in which the space is measured in linear feet produces one set of patterns. Final space allocations in which the space is measured in square feet produces another set of patterns. By this way, each combination resulting from (i) final space allocations under different formats (ii) delta space allocations under different formats and (iii) space allocation logic under different format produces a set of patterns individually Each set of patterns may be termed individually as ‘basic patterns’ or ‘basic set of patterns.’
Mismatch score—Once the set of patterns are generated, at step 314 of the method 300, the one or more hardware processors 104 generate a store level mismatch score by comparing existing floor plans with the set of patterns derived from each of the final space matrix, the delta space matrix, and the space allocation order matrix.
At step 316 of the method 300, the one or more hardware processors 104 perform three level of iterations in which first level iteration is performed to select a set of pattern from the patterns derived from the final space, delta space and allocation order and second level iteration is performed to ensure its fitment into the aisle of retail floors and third level iteration is done to improve quality of the patterns further. First level iteration uses predefined optimization rules and second level iteration, and third level iteration use dynamic optimization rules in real time. The three level iterations comprising:
Locating the deviating stores—Euclidean distance is calculated between column vector of a store and pattern followed by the store and it is repeated for all the stores following a pattern. Here the column vector of a store represents the vector which was used as input for pattern formation. The most deviating store is identified by the following equation
Deviating_Store=max(Euclidean distance between column vector of the store and the pattern) (1)
Based on equation (1), deviating store is identified for any of the format of the final space, the delta space, and the space allocation order. In case of pattern derived from final space under different format, deviating category within the deviating store is identified using below equation
Deviating_category=max(qi−pi)2 (2)
As per equation (2), deviating category for the deviating store is identified and used to select the category that require dynamic optimization rules. Thus, identification of deviating stores and deviating categories enable to choose the stores and categories that require fine tuning to improve quality of patterns. Thus, only part of inputs is modified or adjusted instead of applying dynamic rules across all the categories or across all the stores. Thus, the identification of deviating stores reduces overall running time of iterations.
If the first and second level of iterations give set of patterns from final space matrix, then quality of those patterns are further improved by third level of iterations using the outcome of patterns of previous iteration within third level iterations as follows.
Identification of sub ranges of space change within ideal range of space change for each category for each store—Each pattern will have value for each category. The value may be final space allocations under different formats The value of each category of a pattern is used by the system to decide the new sub range of space change for a category. The value of each category of the pattern becomes as the center of new sub range and the center value is compared with upper bound and lower bound of ideal range of space change of a category or the upper bound and lower bound of sub range of space change of a category of previous iteration and the nearest one is chosen as new sub range of space change for a category. The difference between the center value and nearest boundary value act is the distance that goes on both direction from the center value.
The range will have lesser interval as compared to ideal range of space and thus leading to quicker outcome during optimization. It is understood that optimization results vary in terms of space recommendations of categories as well as running time based on the range used for each category. If the range is higher, it will have more iterations as compared to lesser range in which number of iterations are lesser and running time will be lower. It enables to stabilize the value of category in the pattern in quicker span of time.
Predefined Optimization Rules Vs Dynamic Optimization Rules.
Predefined optimization rules—. Initially the basic patterns are formed using predefined rules. The predefined rules may be same for across categories and across stores. Alternatively, the predefined rules may vary across categories for a store. System may decide those rules based on historical space variations for the categories. They are termed as ‘ideal range of space change’ for a category and it is also categorized under predefined rules.
Dynamic optimization rules—The rules are decided by system based on the outcome of each iteration and adjusting of rules is done based on the outcome. In an instance, if one store is deviating from the pattern slightly, then the rules for the deviating store is adjusted towards narrowing the difference between store and the pattern and pattern is recreated using the new outcomes provided by the adjusted rule to validate for the deviation of the store. As pattern is formed from inputs from many stores, the procedure needs to be performed in an iterative way by adjusting the rules as per latest outcome. In another instance, one store may be differing from the pattern in terms of fitment into the aisle slightly, then the rules for the deviating store is adjusted towards narrowing the mismatch error between aisle of the store and the pattern. As pattern is formed from inputs from many stores, the procedure needs to be performed in an iterative way by adjusting the rules as per latest outcome. Thus, the rule is termed under dynamic optimization rules. The identification of difference between a store and the pattern may be identified in many ways within the scope of present disclosure. As an example, one approach was disclosed using Euclidean distance. The rule adjustment to narrow down the difference or mismatch score may be done in many ways within the scope of present disclosure.
At step 318 of the method 300, the one or more hardware processors 104 generate a set of floorplans and planograms in accordance with the set of patterns received from third iterations. The floorplans and planograms are recommended to each of the plurality of stores in accordance with the set of patterns received from third iterations. The set of floorplans and planograms are generated based on the set of patterns received from third iterations. The floorplans and planograms are recommended to each of the stores based on the set of patterns received from third iterations. All the stores followed or associated with a pattern are assigned with a floorplan. For example, as depicted in
It becomes challenging to carry out floor planning execution at individual store wise. All the stores followed or associated with a pattern are assigned with a floorplan and thus minimization of effort involved in floor planning execution is achieved. The effort minimization is directly related to quality of patterns. Quality of the set of patterns is determined from variance contribution of the set of patterns of PCA.
All the stores followed or associated with a pattern are assigned with a planogram. For example, as depicted in
During remodeling of stores, retailers need guidelines to choose the categories that need to be modified drastically and the order of allocation in pattern format gives great insight in deciding the category selection for remodeling in two ways comprising, category selection for remodeling and store selection for remodeling:
As each pattern has information of prioritization of categories and number of stores adhering those patterns, it enables to decide the number of top categories that could be selected for remodeling of stores based on the overall budget allocated for remodeling. Thus, using the pattern, it improves overall implementation efficiency as it enables simultaneous selection of categories for remodeling and implementations for many stores by reducing overall cost and associated factors.
Reasons (set of reasons) that lead to the formation of the set of patterns received from third iterations are identified based on sales drivers associated with one or more stores. Canonical discriminant analysis is carried out between patterns tagged to each store and store sales drivers. Sales drivers are those factors which are captured at store level and influence sales of the items of the store. Few examples of sales drivers are demography, weather, etc. Sales drivers for a predefined period for a store is processed to bring in a suitable format to map with pattern ID tagged to that store. Canonical discriminant analysis is a multivariate technique, and it is concerned with determination of a linear combination of sales drivers such that the differentiation between patterns is maximum. Canonical analysis attempts to mine all possible linear combinations of sales drivers and provides top combinations in the form of canonical variable and noted as ‘Canonical variable X’ such that the differentiation between patterns is maximum. It enables to apply the canonical variable X to predict the expected pattern which is used for many retail strategies. In one instance when a retailer tries to open a new store in a new location during next season, canonical discriminant model is able to predict potential categories that suit for the new location. It is derived from the predicted pattern by the canonical discriminant model for the given sales drivers that is present in the trade area of new location and forecasted weather conditions for the next season.
Visualization of space allocation mechanism for a store through animation videos is enabled by using corresponding “cumulative total allocation order vectors’ as displayed in
Odd iteration/event=max(Euclidean distance of iterations between two stores) (3)
Similarly, top few iterations could be selected based on magnitude of Euclidean distance. The identified iteration ids are noted in ‘matrix with cumulative totals’, filtered and used for animation
Odd category=max(qi−pi)2 (4)
The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g., any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g., hardware means like e.g., an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g., an ASIC and an FPGA, or at least one microprocessor and at least one memory with software processing components located therein. Thus, the means can include both hardware means, and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g., using a plurality of CPUs.
The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various components described herein may be implemented in other components or combinations of other components. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 202221042736 | Jul 2022 | IN | national |
