Systems and methods for allocating categories of products to a designated space

Abstract

Disclosed herein are systems, methods and a computer readable storage medium for allocating products to a designated space. Preferably the designated space is a space in a display and / or storage environment, such as a physical retail environment, virtual retail environment or a space of a distribution centre.

Description

SYSTEMS AND METHODS FOR ALLOCATING PRODUCTS TO A DESIGNATED SPACEFIELD OF THE INVENTION

[0001] The present invention generally relates systems, methods and a computer readable storage medium for allocating products to a designated space, preferably being a designated space in a display and / or storage environment, such as a physical retail environment, virtual retail environment or a space of a distribution centre.BACKGROUND TO THE INVENTION

[0002] Physical retail store space is a critical asset for success for most traditional retailers. The direct cost of store space is typically one of the top 3 costs retailers bear and the way in which store space is used has significant influence over the other two most significant retailer costs being: (1) costs of inventory; and (2) store labour costs.

[0003] Optimally allocating available merchandisable floor and fixture space to product categories offers significant opportunities to enhance customer engagement with a retailer’s brand and their shopping environments and to encourage increased sales and margins while supporting operational efficiency.

[0004] However, large multi-store and multi-category retail business currently face significant challenges when trying to determine the best use of space in each store to: (1) meet the needs of distinct populations of local customers; and (2) deliver optimal sales, margins, operating costs and deliver other operational objectives. There are a diverse range of factors influencing the overall and category-specific performance of merchandisable floor and fixture spaces in each store. These include: store specific customer preferences, for instance demographic, affluence, life stage and shopping mission drivers; demand patterns or characteristics of the product category; space-related commercial performance characteristics of each product category and the corresponding other product categories competing for space in the store; the physical environment in each store - particularly the available space to beSUBSTITUTE SHEET (RULE 26)merchandised; variations in the marginal value of incremental space allocations to different categories; and other factors such as the retailer’s growth strategies and brand positioning.

[0005] Retail space allocation optimisation methodologies are currently very underdeveloped with retail floor planning teams often reliant on unsophisticated ‘rules-of-thumb’ that results in similar allocations of space across all stores in the retailer store network regardless of these location-specific performance variations.

[0006] Often, assessments of spatial performance indicators for different product groupings are based on out-dated averages that don’t reflect the marginal value of various space allocation choices.

[0007] Retail floor planning teams generally lack the skills to evaluate and quantify the impact of the extensive number of external factors that influence product category spatial performance in each store. In addition, the current methods to determine optimal space allocation solutions for multiple stores in large retailer store networks fail to take into account all the diverse factors that are recognised as being influential to space allocation that both meets the demand of local customers and optimal operational directives.

[0008] In most multi-category retail businesses, the number of valid product category space allocations available that will fill the available merchandisable space in a single store is many orders of magnitude greater than the capacity of traditional analysis tools such as spreadsheet applications to process to find the optimal solution. Efficiently generating optimal space allocation solutions for the hundreds or thousands of stores in many large retailer store networks is also well beyond the capacity of these business tools.

[0009] To efficiently and accurately solve the product allocation in a retail space optimisation problem is to give a globally optimal solution for each store taking into account all the wide variation in each of the multiple influential factors has been too computationally expensive to perform efficiently or is not able to be calculated using the blunt business tools commonly available. To solve this optimisation problem is beyond the capability of the central planning teams within retailer organisations using standard business analysis tools.SUBSTITUTE SHEET (RULE 26)Complicating the computation of this problem is the fact that there is often significant variation in the factors that influence this optimisation problem that has not been satisfactorily addressed by either any earlier computational techniques or the current business tools available to the retail floor planning teams.

[0010] It is desirable for embodiments of the present invention to address or at least ameliorate one or more of the disadvantages of the methods or systems above.

[0011] It is to be understood that, if any prior art publication is referred to herein, such reference does not constitute an admission that the publication forms a part of the common general knowledge in the art, in Australia or any other country.SUMMARY OF THE INVENTION

[0012] According to an aspect of the present invention there is provided a product allocation method for performance by a computer having a processor and a memory, the method comprising: storing, in a database in communication with the computer, information in association with respective identifiers of spaces of a network of spaces of a product display and / or storage environment for displaying and / or storing a plurality of products; receiving, from a user device in communication with the computer over a computer network, a user input comprising a designated space, at least one space performance measure and product data corresponding to a sub-set of products; in response to receiving the product data, automatically retrieving information from the database and automatically determining a correspondence of the plurality of products for display and / or storage in the network of spaces with the sub-set of products in order to automatically identify at least one group of spaces from the network of spaces; in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure;SUBSTITUTE SHEET (RULE 26)mapping, automatically, the determined assessment of the spaces in the at least one group to an assessment in the designated space; determining, automatically, a ranking of a plurality of product allocations of the designated space from the mapped assessment; determining, automatically, using the computer, allocation of products of the sub-set of products in the designated space of the product display and / or storage environment on the basis of the determined ranking. displaying 2-D geographic map of how each product of the sub-set of products is allocated to the designated space, upon receiving a user input.

[0012] According to an embodiment of the present invention, the method includes, determining a tabulation of how each product of the sub-set of products is allocated to the designated space. A tabulation and / or 2-D geographic map assists the team to easily show the allocation of space and plan accordingly.

[0013] According to an embodiment, the step of automatically generating the space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure includes performing normalisation of values of the at least one performance measure for reduction of variation within the at least one group of spaces.

[0014] Preferably, performing normalisation of values by one or more of the following: adjustment of sales to reflect constant customer density, the at least one performance measure, internal factors which impact the at least one performance measure, external factors which impact the at least one performance measure.

[0015] According to an embodiment, the determined assessment of the spaces in the at least one group to an assessment in the designated space includes: calculating at least one function which fits the relationship between spaces of the at least one group of stores and the at least one performance measure; determining, automatically, a set of space allocation values allocated to the sub-set of products and their associated at least one performance measure from the at least one function;SUBSTITUTE SHEET (RULE 26)calculating a set of allocated space allocation values for each respective product of the subset. Preferably, the at least one performance measure is for at least one product.

[0016] Preferably, the relationship between spaces of the at least one group of stores and the at least one performance measure is calculated on the basis of historical information automatically retrieved from the database. Preferably, the historical information includes the following: space information, product and product category information, information relating to space usage and information relating to performance of product in stores.

[0017] According to an embodiment, the method includes the step of grouping retail stores on the basis of the at least one performance measure includes building a hierarchical clustering model based on a plurality of space performance measures.

[0018] According to an embodiment, the method includes the step of determining the relationship between allocated space and the at least one performance measure is made on the basis of historical data.

[0019] According to an embodiment, calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes calculating a yield curve which determines volume sales performance in relation to variations in space allocation.

[0020] According to an embodiment, calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes calculating a cost curve which determines performance measures in relation to the at least one product category. Cost curves can be estimated as mathematical curves or business driven algebraic estimates. The cost module can also be used in rare cases to account for expected future growth / decline in performance driven by external factors.

[0021] Preferably, the at least one yield curve and / or the at least one cost curve include polynomial functions. More preferably, the polynomial functions include one or more of the following: single form algebraic function, spline functions and linearised forms of polynomial functions.SUBSTITUTE SHEET (RULE 26)

[0022] According to an embodiment, calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes a combination of at least one yield curve and at least one cost curve.

[0023] According to an embodiment, wherein calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes determining a plurality of combinations of at least one yield curve and at least one cost curve for each of the last one group of spaces and products of the sub-set of products.

[0024] According to an embodiment, the method includes the step of determining which of the plurality of combinations have a designated value of the at least one performance measure for allocation of the products of the sub-set of products to designated space.

[0025] According to an embodiment, the method includes the step of determining which of the determined combinations have a best fit to the relationship between allocation of space for the subset of products and the at least one performance measure comprises a statistical test. Preferably, the statistical test includes one or more of the following: mean squared error, sum of squared error, bias.

[0026] According to an embodiment, the at least performance measure includes product sales, product margin, product sales, factors associated with product margin, factors associated with cost, inventory cost, store operational costs, supply chain operational costs and product wastage costs. Preferably, the at least one performance measure is a performance measure for at least one product.

[0027] According to an embodiment, the method includes the step of determining which combinations have a pre-selected value of the at least one performance measure for optimal allocation of the products of the at least one product category to the space.

[0028] According to an embodiment, the step of determining which combinations have a pre-selected value of the at least one performance measure comprises use of the Knapsack algorithm. Preferably, the Knapsack algorithm comprises the following:SUBSTITUTE SHEET (RULE 26)

[0029] where each product has weight (w) and the weighted sum is the size of the retail store,x e {0,l} i = l ... k, j e Ni

[0030] According to an embodiment, the method includes the step of calculating a list of the sub-set of products with their associated allocated space values that optimise the at least one performance measure using the combinations having a pre-selected value of the at least one performance measure. Preferably, the step of calculating the list of products with their retail space values that maximise the selected at least one performance measure comprises use of a greedy algorithm.

[0031] According to an embodiment, if the step of calculating the list of products with their retail space values results in unallocated space, then the step includes allocating products to the said unallocated space. Preferably, the step of allocating products to the said unallocated space comprises use of the 0 1 knapsack algorithm.

[0032] According to an embodiment, wherein performing normalisation of values of the at least one performance measure for reduction of variation within the at least one group of spaces includes normalising by one or more of the following: adjustment of sales to reflect constant customer density, the at least one performance measure, internal factors which impactSUBSTITUTE SHEET (RULE 26)the at least one performance measure, external factors which impact the at least one performance measure.

[0033] According to an embodiment, the at least one yield curve and / or the at least one cost curve includes one or more of the following: single form algebraic function, spline functions and linearised forms of polynomial functions.

[0034] According to an embodiment, the at least performance measure includes product sales, product margin, product sales, factors associated with product margin, factors associated with cost, inventory cost, store operational costs, supply chain operational costs and product wastage costs.

[0035] According to an embodiment, the statistical test includes one or more of the following: mean squared error, sum of squared error, bias, mean, median.

[0036] According to an embodiment, the method includes determining a plurality of steps for allocating products to the designated space of the space or portion of space

[0037] According to an embodiment, the method includes the step of determining an optimal size of a retail store or portion of said store.

[0038] According to an embodiment, the method includes determining an optimal size of a retail store or portion of said store includes combining said plurality of steps for allocating retail space to a retail store with retail space cost estimates.

[0039] According to an embodiment, the method includes, in response to responses from one or more user devices, receiving user inputs comprising the following: space parameters; optimisation parameters; minimum or maximum space sizes; spaces / jumps between allocated products; product treatment; custom space.

[0040] According to an embodiment, the user device is one of a plurality of user devices.SUBSTITUTE SHEET (RULE 26)

[0041] According to another aspect of the present invention there is provided a product allocation system for performance, the system comprising: a computer having a processor and a memory; a database in communication with the computer, the database being adapted to store information in association with respective identifiers of spaces of a network of spaces of a product display and / or storage environment for displaying and / or storing a plurality of products; wherein the processor is adapted to execute a plurality of modules stored in the memory, each module being able to execute a set of instructions, and wherein the modules comprise: a spatial performance assessment module adapted to: in response to receiving the product data, automatically retrieving information from the database and automatically determining a correspondence of the plurality of products for display and / or storage in the network of spaces with the sub-set of products in order to automatically identify at least one group of spaces from the network of spaces; in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure; in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure; a product-space performance module adapted to: mapping, automatically, the determined assessment of the spaces in the at least one group to an assessment in the designated space; determine, automatically, a ranking of a plurality of product allocations of the designated space from the mapped assessment; a space optimisation module adapted to: determining, automatically, using the computer, allocation of products of the sub-set of products in the designated space of the product display and / or storage environment on the basis of the determined ranking.SUBSTITUTE SHEET (RULE 26)

[0042] According to an embodiment of the present invention, the system includes, in response to receiving, from a user device in communication with the computer over the computer network, a user input, determining a tabulation of how each product of the sub-set of products is allocated to the designated space.

[0043] According to an embodiment of the present invention, the system includes a display module adapted to: in response to receiving, from a user device in communication with the computer over the computer network, a user input, determining a 2-D geographic map of how each product of the sub-set of products is allocated to the designated space; displaying, on the user device, the 2-D geographic map.

[0044] According to yet another aspect of the present invention there is provided a computer readable storage medium that stores instructions which, when executed by one or more processors of a computer, cause the computer to run instructions stored in a memory of the computer: storing, in a database in communication with the computer, information in association with respective identifiers of spaces of a network of spaces of a product display and / or storage environment for displaying and / or storing a plurality of products; receiving, from a user device in communication with the computer over a computer network, a user input comprising a designated space, at least one space performance measure and product data corresponding to a sub-set of products; in response to receiving the product data, automatically retrieving information from the database and automatically determining a correspondence of the plurality of products for display and / or storage in the network of spaces with the sub-set of products in order to automatically identify at least one group of spaces from the network of spaces; in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure; mapping, automatically, the determined assessment of the spaces in the at least one group to an assessment in the designated space; determining, automatically, a ranking of aSUBSTITUTE SHEET (RULE 26)plurality of product allocations of the designated space from the mapped assessment; determining, automatically, using the computer, allocation of products of the subset of products in the designated space of the product display and / or storage environment on the basis of the determined ranking.

[0045] In an embodiment of the present invention, the computer readable storage medium includes instructions for determining a tabulation of how each product of the sub-set of products is allocated to the designated space.

[0046] In an embodiment of the present invention, the computer readable storage medium includes instructions for displaying 2-D geographic map of how each product of the sub-set of products is allocated to the designated space, upon receiving a user input.

[0047] A system, method or a computer readable storage medium as described above wherein the space is a retail store of a physical retail environment, virtual retail environment or a space of a distribution centre.BRIEF DESCRIPTION OF THE DRAWINGS

[0048] One or more embodiments of the present invention will hereinafter be described with reference to the accompanying Figures, in which:

[0049] Fig. 1 is a schematic diagram of an example network environment in which the present invention may be implemented;

[0050] Fig. 2 is a flowchart of a retail space allocation method according to preferred embodiments of the present invention; and

[0051] Fig. 3 is a flowchart of an exemplary optimisation routine for use in the retail space allocation method of Fig. 2.DETAILED DESCRIPTION OF THE DRAWINGS

[0052] FIG. 1 illustrates a computing environment 10 in which aspects of the present invention are implemented. The environment 10 is a networked environment comprising a server system 100 in communication with a user system 20 over one or more communicationSUBSTITUTE SHEET (RULE 26)networks, such as the internet 30. Aspects of the computer processing described below are performed by a server application 100 (hereinafter referred to as an “item allocating method”) executing on the server system 100 and a user application executing on the user system 20.

[0053] The server system 100 further includes a data storage (not shown) on which data managed by the retail space management system 100 is stored. The data storage is typically a storage medium such as a hard drive (or collection of hard drives). A database management system 102 executing on server system 100 implements a database on the data storage for storing and retrieving data managed by the retail space management system 100. A user, for example, a member of the retail store planning team, can establish an account with retail space management system 100 and store data pertaining to the user preferences and retail store preferences. The totality of user accounts established with the retail space management system 100 are stored in data storage.

[0054] Although throughout this disclosure, while reference is made to space or retail space, the following described system and methods are equally applicable to similar physical and digital environment multiple product categories must be allocated shares of space within a constrained total space. Examples include online retail stores, storage spaces and supply chain distribution centres where each retail store, storage space or supply chain distribution centres can be one entity of multiple entities in a network of a product display and / or storage environment.

[0055] The server system 100 has been illustrated as a single system. The server system 100 can, however, be a scalable server system comprising multiple nodes which can be commissioned / decommissioned based on processing demands. Typically, server systems are server computers that provide greater resources (e.g. processing, memory, network bandwidth) in comparison to user systems.

[0056] A database management system 102 may access data storage as part of the server system 100. However, the data storage could be a separate system in operative networked communication with the server system 100. For example, the data storage could be a networked-attached storage device, an entirely separate storage system accessed via a database management system, or any other appropriate data storage mechanism.SUBSTITUTE SHEET (RULE 26)

[0057] As described in further detail below, the retail space management system 90 performs various operations in response to commands received from (and initiated at) user application 112. As such, when executed by the server system 102, the retail space management system 100 configures the server system 100 to provide server-side functionality to user application 112. To provide this functionality, the retail space management system 100 comprises one or more suitable application programs, libraries, or other software infrastructure.

[0058] Where the user application 112 is a web browser, the retail space management system 100 will typically be, or interact with, a web server such as a server implemented with the node.js runtime environment. Where the user application 112 is a dedicated application provided specifically to interact with retail space management system 100, the retail space management system 100 will typically be, or interact with, an application server. Server system 100 may be provided with both web server and application server applications 90 to enable it to serve both web browser and dedicated user applications.

[0059] The server system 100 and user system 20 communicate data between each other either directly or indirectly through one or more communications networks. The communications network may comprise a local area network (LAN), a public network (such as the Internet 30), or a combination of networks.

[0060] The environment 10 is a networked environment. In alternative embodiments, however, the retail space management system 100 may be locally installed on a computer system (e.g. user system 20) and adapted to perform operations for that user system 20 alone. In this case a server system is not required.

[0061] While only one user system 20 is depicted in environment 10, a typical environment would typically include many more user systems served by the server system 100.

[0062] While user system 20 can be any type of computer system, including a desktop computer or laptop computer 80, it may commonly be a smartphone or a tablet device. When executed by the user system 20, the user application configures the user system 20 to provide user-side functionality and interact with server system 100 (or, more specifically, the retail space management system 100 running thereon).SUBSTITUTE SHEET (RULE 26)

[0063] The user application may be a general web browser application (such as Chrome, Edge, Safari or the like) which accesses the retail space management system 100 via an appropriate uniform resource locator (URL) and communicates with retail space management system 100 via general world-wide-web protocols (e.g. http, https, ftp) and application programming interfaces 90 (APIs) (e.g. REST APIs). Alternatively, the user application may be a specific application programmed to communicate with the retail space management system 100 using defined API calls.

[0064] A given user system 20 may have more than one user application installed thereon, for example both a general web browser application and a dedicated programmatic user application.

[0065] The server system, as illustrated in Fig. 1, in addition to the application programming interface and user interface 90 and data management system 102, also comprises a retail space configuration system 106 and retail space optimisation system 104. Any reference to space is a reference to retail space throughout this disclosure and which has been shortened for the purposes of brevity and ease of use.

[0066] The data management system 102 can comprise a data preparation and enrichment module 104 and at least one database storing retail store data 110. The data preparation and enrichment module 104 translates business’s retail store data, data relating to products and product category into data to be used and then stored in storage as retail store data 110, validates and cleans data, and provides lineage and auditability of data transformations. The retail store data 110 contains historical raw data on how retail space is used across retail stores of a retail store network and product category, historical (raw) data on performance of product categories in stores, raw data containing list of retail stores in the store network, product categories, products. The retail store data 110 also contains processed data which has been processed by the data preparation and enrichment module 104 in preparation for use in other modules, including those in the space optimisation system 104 and space configuration system 106. The data preparation can also include rolling up the involved performance measure information for a time period. The data preparation and enrichment can include mapping of financial product hierarchy to merchandise range driven product hierarchy, mapping at least one performance measure (KPI) data to allocation space data as described within this disclosure.SUBSTITUTE SHEET (RULE 26)

[0067] The retail space configuration system 106 has a retail space configuration module 106, as illustrated in Fig. 1, which has a retail space configuration module 170 and at least one database containing performance data 180 and configuration data 190. The database containing performance data 180 and configuration data 190 can be the same or different to that of the database containing retail data 110. The retail space configuration module 170 is adapted to receive responses from the user device 80 or user system 85 of the user system 20 relating to selection of configuration parameters used in the retail space optimisation system 104 and space configuration system 106 and to store the configuration parameters in configuration data 190. For example, the configuration parameters can include user selected parameters relating to each product, each product category and retail store information. The retail space configuration system 85 is also adapted to store the afore-mentioned user responses in a database, which may be the same database as that used to store data in retail store 110 or a different database.

[0068] In a particularly preferred embodiment, the user can select or nominate from the following different configuration parameters which can be set for each product category and store: whether the product category or retail store is included or not; optimisation zone; minimum retail space available of the retail store, product or product category; maximum retail space available of the retail store, product or product category; spacing, for example between products and / or product category, shelves etc,; product treatment; and custom spacing. These configuration parameters can be provided at multiple levels of hierarchy (optimisation zone / store / store cluster / all stores) or (product, product category) or portion of retail space (shelf, row, department, section / optimisation zone).

[0069] The optimization zone comprises an area where the products in that area are optimised for a specific store, thereby comprises a sub-zone of a store, or a sub-store. For example, the optimisation zone may relate to chilled products, and thus as only chilled products can be optimised within a chilled space of a store, the chilled space would be an optimisation zone. Similarly, ambient products can be optimised only in the ambient space and thus the ambient space would be another optimisation zone.

[0070] Product treatment relates to whether the product is provided with optimal space or fixed within a specific store. If a product is fixed within a specific store, it can be fixed to its current space or to a custom space provided by the business.SUBSTITUTE SHEET (RULE 26)

[0071] The retail space configuration module 170 is adapted to provide to the user at least one default configuration parameter based on historical (raw) retail space performance data. In one example, for minimum and maximum space, the default can be set at a cluster level by looking at the range of a product's space within a cluster. Product treatment defaults to an optimal setting except those where very limited data is available. The historical retail space performance data is stored in the retail space performance data 180. In addition, the raw retail space performance can be processed by the data preparation and enrichment module 104 and stored in the retail space performance 180. Processing of the data automatically by use of this module 104 assists the efficiency, increase accuracy and use of the data used in the modules following module 104 and thus reduces time and effort required by a user in the system and methods overall.

[0072] The retail space optimisation system 104 contains a number of modules which perform different tasks relating to product allocation of a sub-set of products for a single retail store, otherwise known as the ‘designated’ store, and which can be expressed as an exemplary method which is to be described in detail in the following paragraphs below. The retail space optimisation system 104 has a retail spatial performance module 120 which is adapted generally to provide a selection of key performance indicators (KPIs) relating to space performance by displaying at the user device / system 80, 85 a list of space performance KPIs and is further adapted to receive a user response relating to a selection, of at least one space performance KPI. The retail space performance module 120 is also adapted to combine two or more KPIs or transforming at least one KPI based on the user response thereby allowing customisation of KPIs to create a custom performance metric. The selection of the KPI or custom performance measure / metric, henceforth known as the space performance measure which is to be optimised. For example, the space performance measure can be sales or margin of a product or product category. The space performance measure selected can include aspects of sales (in dollar terms and in magnitude), margin and cost factors that include (but are not limited to) inventory cost, store and supply chain operational costs, product wastage costs, etc. Custom space performance measures can be devised for a business based on their current and future strategic goals. The space performance measures selected can also include aspects of sales (in dollar terms and in magnitude), margin and cost factors that include (but are not limited to) inventory cost, store and supply chain operational costs, wastage, product wastageSUBSTITUTE SHEET (RULE 26)costs, etc, factors which relate to from margin to profit, volume, units, and combinations or translations of any of those afore-mentioned.

[0073] The retail spatial performance module 120 is also adapted to determine a hierarchical cluster model based on at least one performance measure, whether user-selected or user-customised as discussed in the previous paragraph. Retail space for a product (or item category - the logical level at which floor or fixture space allocations are decided) do not change often in a retail store and therefore trying to understand impact of differing space allocations on product performance is a challenge. To address this challenge, stores in the retailer’s network that are similar can be grouped or ‘clustered’ together with retail spatial performance assessed across these similar stores with particular reference to the impact of differing allocations of space to a given product or product category, and particularly relating to those stores which display and / or store products which correspond to the sub-set of products for allocation in the designated retail store.

[0074] In a particularly preferred embodiment, the retail spatial performance module 120 is adapted to determine which stores of the list of retail stores are to be grouped based on the at least one selected performance measure, and / or according to which stores have allocated space to the sub-set of products which will be displayed and / or stored in the designated retail store or space. The clustering of stores is based on store performance attributes that are relevant to the business’s current and future strategic goals and are directly applicable to the optimal allocation of product categories across the available floor or fixture space in each store. The data used by the retail spatial performance module 120 for the clustering analysis can be raw data stored in the retail data 110, performance data 180 or configuration data 190 or any of the afore-mentioned data processed by the data preparation and enrichment module 104.

[0075] The retail spatial performance module 120 is also adapted to group the determined stores in a hierarchy tree model based on one or more of the performance measure(s), also known as a hierarchical clustering model or analysis. Preferably, a multi-dimensional hierarchical clustering analysis is performed by the retail spatial performance module 120. Some examples include sales velocity, department sales penetration, costs-to-serve optimisation etc. The hierarchy clustering analysis also ensures breadth of store sizes within each cluster which in turn enables assessment of the retail space performance outcomes generated from a variety of retail space allocation values for a given product or productSUBSTITUTE SHEET (RULE 26)category of the sub-set of products which is to be discussed in the following paragraphs. The algorithm used in the clustering analysis can be any of those used in commercial analysis packages such as MATLAB, MATHEMATIC A or any open source implementations such as Octave. When a combination of performance metrics needs to be used with varied weights, a custom multi-target regression tree algorithm is used to enable meeting retailer specific strategic needs. The custom algorithm enables use of performance metrics in it’s raw form (and not normalised and indexed like typical out of the box solutions). The algorithm also helps build the tree with features input by the user to make it more business relevant as needed.

[0076] The retail spatial performance module 120 is also adapted to provide reports on statistical analysis performed on each clustering model for determining the usefulness and accuracy of the model. The statistical analysis includes measures such as square error value, within cluster inertia, inertia gain and other like measures.

[0077] By way of the spatial performance module 120, the retail stores of the store network are organised into strategic space allocation clusters or ‘groups’, variation of space allocations for product categories within each cluster is advantageously achieved. The store clusters are also expected to normalise many other internal and external factors that may have an impact on product performance. These assertions therefore enable a system where impact of space on product performance can be ascertained with high levels of accuracy.

[0078] The space optimisation system 106 also has a space performance normalisation module 125 which is adapted to normalise the historical retail space performance data to advantageously account for store-level variation in the total volume of retail store performance. This normalisation is essential because of store-level variation in total volume of performance within each cluster for stores of a given size. The space performance normalisation module 125 is adapted to instruct the server system 100 to display a list of normalisation technique to the user and to receive a user response selecting at least one of the normalisation techniques. Preferably, the space performance normalisation module 125 is adapted to determine a ranking of the normalisation techniques based on the at least one pre-selected performance measure and to default use of the top-ranked normalisation technique. This normalisation module 125 assists to automate the process for accounting for store-level variation which greatly increases the accuracy and efficiency by which the method and system calculates the product allocation in a given space, and also reduces effort to the user.SUBSTITUTE SHEET (RULE 26)

[0079] The space optimisation system 106 also has a hierarchy mapping module 130 which is adapted to translates hierarchy of historical (whether raw or processed) data to a forward-looking hierarchy. The forward-looking hierarchy can include forwarding mapping which is based on mapping changes provided as an input. Users can customise the hierarchy in which the results are provided. This can be different to the hierarchy available by use of historical data. A classic example of this is when there is a change in product hierarchy. If a set of stock-keeping units, (SKUs), otherwise known as the identifiers that defines a product at the identifiable inventory level, i.e. size, style, colour, is mapped to product A today (historically) but will be split as B and C in the future, the algorithm can either perform the future looking mapping to historical data and perform optimisation for B and C or it can perform optimisation for A and split as B and C at the output stage.

[0080] The space optimisation system 106 has a performance curve module 160 which is adapted to assist the understanding the relationship between retail space and performance using historical data by describing these relationships as curve functions; this can be for a particular product or product category. This leads to two types of functions: yield curves that estimate pure volume sales-led performance in relation to different allocations of space; and cost curves that estimate various retail space allocation driven or sales driven cost elements associated with the product / product category.

[0081] Yield curves, in this performance curve module 160, are expected to follow certain rules:Positive relationship: A positive allocation of space leads to positive performance. No retail space allocation for a given product or product category implies no performance hence the curves should start at (0,0).Increasing function: Increased allocation of retail space, per product / product category, typically results in higher performance.Diminishing returns to additional space.

[0082] These yield functions can take any form as long as they satisfy the above conditions. These functions include single form algebraic function, spline functions that behave differently with increase in space, linearised forms of polynomial functions and any similar types of functions that a person skilled in the art would reasonably find to be suitable. As such a library of yield functions can be accumulated and stored, for example in system library 165.SUBSTITUTE SHEET (RULE 26)

[0083] Cost curves are expected to behave similarly to the yield curves but in the opposite direction in relation to retail space and changes to space, and with much lower magnitude of impact. A repository of such cost curves can also be accumulated for various cost elements keeping in mind the rules and functional form as described for yield curves, and also stored in a database library, for example system library 165.

[0084] The performance curve module 150 is also adapted to perform curve fitting of the yield and cost curves. Given that each cluster product combination results in a unique yield curve and there are multiple curves to choose from, the number of curves that need assessment can comprise thousands of curves.

[0085] The performance curve module 150 is adapted to display a pre-defined list of types of yield curves and cost curves to fit to the user via the user application 20 on the user device or system 80, 85, and to receive the user response and store that response in a database, for example configuration data 190. Alternatively, the performance curve module 150 is adapted to provide a display to the user, via user system 20, of a determination of a ‘best fit’ type of yield curve and cost curve to fit for every cluster and product pair combination, based on historical data and / or selected criteria. The performance curve module 150 is adapted to determine the ‘best fit’ curve, based on an automated process, for every cluster product pair, given the magnitude of the number of curves from which to choose. This performance curve module 150 assists to automate the process for providing a ‘best-fit’ curve which greatly increases the accuracy and efficiency by which the method and system calculates the product allocation in a given space, and also reduces effort to the user, who may not have the necessary skills to select an appropriate curve. In a preferred embodiment, the ‘best-fit’ curve is one which is a positive, increasing and saturating curve. Therefore, one of the criteria that can be selected by a user for a yield curve includes a positive, increasing and saturating curve. A cost curve on the other hand can take any functional form that is positive and increasing. It can also be a simple algebraic function as described in 0021. In another embodiment, the criteria selectable by a user can include a valid minimum, i.e. where the curve intersects the x-axis. Other criteria selectable by a user can include accuracy metrics as discussed in the following paragraph.

[0086] This ‘best-fit’ selection is based on an approach that considers various statistical tests that include but are not limited to mean squared error, sum of squared error, bias, etc.SUBSTITUTE SHEET (RULE 26)While these statistical tests assess the accuracy of the curves within the data range, the behaviour of the curves, both yield curves and cost curves, beyond their range is also considered to ensure things like no negative KPI at low values and tapered growth at large values.

[0087] The performance curve module 150 is adapted to receive a user response which accepts the ‘best fit’ determination as displayed or a user response which selects another type of curve to fit. The performance curve module 150 is adapted to display to the user, via the user system 20, information on the different types of curves for review before or during selection. The performance curve module 150 is adapted to provide to the user via the user system 20 a list of curve types by leveraging information from libraries and APIs and to allow a user response associated with a selection of at least one of those curve types and storing that information for use in configuration data 90.

[0088] Setting Category Minimum and Maximum Space Allocation Constraints. In one example, the user response can include selection of allocation constraints to assist in selecting curves, for example constraints can include minimum and maximum allocated space. In an example the default can be set at a cluster level by looking at the range of a product's space within a cluster. In another example, product treatment defaults to an optimal setting except those where very limited data is available. For each product, a range of acceptable space values is further determined based on: existing space across the cluster and store network, strategic business goals, as well as the accuracy and validity of the yield function.

[0089] The space optimisation system 106 also includes a product-space performance module 140 which is adapted to determine at least one yield curve, or at least one cost curve or a combination of at least one yield curve and at least cost curves for assisting prediction of performance KPIs for every possible retail space per product / product category. The productspace performance module 140 is also adapted to calculate values of space performance measure (KPI), by way of an optimisation engine module 160, as selected by the user for optimisation in configuration module 170 and stored in configuration data 190. The productspace performance module 140 is also adapted to calculate an incremental performance measure (KPI) for every possible retail space for each product / product category. Incremental performance measure (KPI) is the change in KPI with respect to change in space. Finally, the product-space performance module 140 is also adapted to provide a ranking of possible productSUBSTITUTE SHEET (RULE 26)allocation spaces based on incremental performance (KPI) per product / product category thereby creating a list of all possible allocated space values for each store and product / product category based on configuration parameters provided by the user as selected in the configuration module 170 and stored in configuration data 190.

[0090] In a particular preferred embodiment, the product-space performance module 140 is also adapted to provide a performance curve for a cluster product combination comprising a yield curve combined with various cost curves to deliver a specific estimated performance value for each cluster-product category-space allocation unit. A multi-dimensional curve comprising a yield curve with at least two cost curves is a novel feature of this process that is also serves as the backbone for the optimisation process which is described in more detail in the above paragraphs. At least two cost curves can be established to reflect either space driven costs or sales driven costs. By combining multiple cost curves, various combinations of sales and space driven costs can be accurately combined to support the calculation of expected total costs and profit at a given combination of space allocation value and sales level.

[0091] The product-space performance module 140 is adapted to input the best-fit performance curves and adjust the performance curves from a cluster level to a retail store level resulting in store product level performance estimation for a given product or product category. Preferably, this adjustment of the performance curves is based on historical data or processed data and configuration parameters as selected by the user. By using the product-space performance curve module 140, each product in the sub-set considered for allocation of retail space in the retail store now will have such a set of space values and their corresponding space performance measures (KPIs). For a fixed designated space / store size, this will lead to a significantly large solution space that cannot be computed in linear time.

[0092] The space optimisation system 106 also includes an optimisation engine module 160. The optimisation engine module 160 is adapted to reduce this large solution space, to reduce the number of data points which might not be needed, by the use of dominance relations. The optimisation engine module 160 is also adapted to optimise the allocation of retail store space by use of algorithms, for example the MILP greedy algorithm and 01 Knapsack algorithms. Finally, the optimisation engine module 160 is adapted to provide a report on each of the constraints - e.g. is any store overallocated (more than the store space). The optimisation engine module 160 is adapted to output results by way of a set of product space allocationSUBSTITUTE SHEET (RULE 26)values for a product / product category along with their corresponding estimated performance measure (KPI). These retail space allocation values can go from a minimum to a maximum space allocation value in sequence or take disjoint values in their space range. The optimisation engine module 160 is adapted to determine the best combination of products in the sub-set and their corresponding retail space allocation that will maximise the user-selected at least one performance measure (KPI), preferably by using an optimisation algorithm or function, while being constrained by their individual minimum and maximum space allowed plus the total amount of space available at the store level.

[0093] In a preferred embodiment, the optimisation engine module 160 is adapted to determine the best combination of products and their corresponding retail space allocation by way of an optimisation function, for example an algorithm that uses the principles of multi choice knapsack algorithm and a 0-1 knapsack algorithm for fringe allocation has been devised to optimise the performance curve with constraints on space at a product level and at an overall level. This normalisation module 160 automates the process for determining the best combination of products and their corresponding retail space allocation which greatly increases the accuracy and efficiency by which the method and system calculates the product allocation in a given space, and also reduces effort to the user, especially if the user does not have the necessary analytic skill to use the optimisation module 160 by way of selecting the appropriate constraints necessary to optimise the performance curve.

[0094] The 0- 1 Knapsack problem is the problem of choosing a subset of products with a fixed volume of space (v) such that the corresponding performance metric (p) is optimised without exceeding the size of the store (S)Xj£ {0,1}, j = 1 ... n

[0095] A generalisation of the Knapsack algorithm that applies to the retail space optimisation problem is the Multi Choice Knapsack problem (MCKP). In MCKP, each product is given a weight(w) and the weighted sum is expected to add up to the size of the retail store.SUBSTITUTE SHEET (RULE 26)The weight is equivalent to the space each product category is expected to take whose sum is constrained by the size of the retail store. The above equation changes to:

[0096] Dominance relations assist the optimisation algorithm to reduce number of data points that might never be needed to be traversed to achieve optimisation. When the dominance relationship is checked on continuous data, Linear Programming (LP) is the preferred mode. Hence the terminology LP un-dominated when used in the MCKP algorithm. In the current situation, LP un-domination translates to no negative values in the performance curve and diminishing function throughout the range of the curve. The database of stored curves takes care of this at an individual level. When curves are combined (say a yield and a cost function), this condition can be violated. Such violating space values are removed before optimisation.

[0097] The optimisation engine module 160 is preferably adapted to use another algorithm, for example a Mixed Integer Linear Programming (MILP) greedy algorithm, on the remaining valid space set to find a list of products along with their space values that maximizes the performance metric while not exceeding the size of the retail store. The algorithm also ensures network level constraints and product distribution constraints are met while deciding the optimal solution. Since an optimal solution is not always guaranteed due to conflicting constraints, there is a possibility of fringe space left over at the end of the optimisation. This fringe space is then allocated using an allocation algorithm, for example the 0 1 knapsack algorithm.

[0098] The optimisation algorithm, for example the greedy algorithm, outputs a set of ordered steps of allocation of retail space for a new or an existing store for sub-set of productsSU BSTITUTE SH EET (RULE 26)or product categories. This set of ordered steps combined with further retail store level per unit of space cost estimates determines the optimal size of a store or a department within the store. This ordered step data is the key ingredient to deliver on the fly scenarios and customisations that include future growth forecasts and other external change factors. The ordered step data also helps achieve optimisation at more granular levels (like a department or even an aisle).

[0099] The optimisation engine module 160 automates the process for determining the best combination of products and their corresponding retail space allocation that will maximise the user-selected at least one performance measure selected. This greatly reduces the effort to the user who may use the optimisation engine module 160 with little or no input yet can optimise the product allocation to easily produce a set of ordered steps of allocating products to the space.

[0100] Given the ordered nature of the ordered step data, this transforms an existing store planogram to the next most profitable planogram change. Such a change helps store planners decide the right order of moving from an existing planogram to the optimal planogram.

[0101] Table 1 below provides a list of the main capabilities as performed by each module 120 to 170 as outlined for claritySUBSTITUTE SHEET (RULE 26)SUBSTITUTE SHEET (RULE 26)

[0102] The retail space allocation method for allocation of space in a retail environment is described as illustrated by reference to Fig. 2 by use of the retail space optimisation system 104 and particularly modules 90 to 190. It is envisaged that the preferred embodiment of this method is directly applicable to ‘brick & mortar’ store-based retail businesses, for example a retailer having greater than 20 stores and greater than 10 product categories to be allocated however other applications are possible as would be reasonably understood by the person skilled in the art, for example online business, or distribution centres.

[0103] The method 200 commences at step 202 where the retail spatial performance module 120 group together similar retail stores on the basis of the at least one performance measure as selected by the user. The retail spatial performance module 120 provides a selection of key performance indicators (KPIs) relating to space performance by displaying at the user device / system 80, 85 a list of space performance measures (KPIs) and is further adapted to receive a user response relating to the selection of at least one space performance measure (KPI) for use in the grouping step. The user determines the at least one space performance measure (KPI) by considering strategic business goals 204 and space drive priorities 206.

[0104] At step 208, the retail spatial performance module 120 determines at least one hierarchical tree models using the selected at least one performance measure using the groupings of like retail stores performed in step 202. At step 210, the retail spatial performance module 120 performs a hierarchical clustering analysis.

[0105] At step 212, the space performance normalisation module 125 normalises the historical retail space performance data within a selected cluster to advantageously account forSUBSTITUTE SHEET (RULE 26)store-level variation in the total volume of retail store performance. The space performance normalisation module 125 instructs the server system 100 to display a list of normalisation technique to the user and to receive a user response selecting at least one of the normalisation techniques which is to be used in step 212. Alternatively, the space performance normalisation module 125 determines a ranking of the normalisation techniques based on the at least one preselected performance measure and to default use of the top-ranked normalisation technique in step 212. The use of default selection where the system and method can pre-determine the best normalisation techniques reduce the effort and time to a user to produce optimal results for product allocation.

[0106] After step 212, the method 200 splits into two sub-methods; sub-method 200a which relates to custom performances measures and sub-method 200b. At steps 214 and 216, the user can create a custom performance metric 214 or cost metric 216 at respective submethods 200a and 200b. The retail spatial performance module 120 provides a selection of key performance indicators (KPIs) relating to space performance by displaying at the user device / system 80, 85 a list of space performance KPIs and receives a user response relating to a selection of at least one space performance KPI for combining two or more KPIs or transforming at least one KPI based on the user response thereby allowing customisation of KPIs to create a custom performance measure or cost measure / metric. Although the default settings can reduce work to a user, the user can also select or customise performance measure easily in a way which was not previously available.

[0107] A third sub-method 200c commences at step 222 where the performance curve module 160 inputs data, including historical data or processed data 224 from the data management system 102, or business knowledge 226 by describing the relationships between retail space and the at least one performance measure as curve functions.

[0108] At step 228, the performance curve module 160 ensures that the curve functions follow certain relationships by performing curve fitting with the following rules (i) positive allocation of space leads to positive performance, i.e. no negative values for retail space and correspondingly no negative values for the selected performance measure (ii) No retail space allocation for a given product or product category implies no performance hence the curves should start at (0,0). (iii) Increasing function: Increased allocation of retail space, per product / product category, typically results in higher performance, (iv) Diminishing returns toSUBSTITUTE SHEET (RULE 26)additional space. The various curve functions can be determined by performance curve module 160 and then stored in system library 165 at step 230.

[0109] At step 218 and step 220, the performance curve module 160 displays at the user device / system 80, 85 a list of curve functions taken from the library 165 at step 230 and receives a user response relating to a selection of at least one curve function for at least one retail space performance measure at step 218 or at least one space cost curve at step 220. Alternatively, the performance curve module 160 automatically determines the best retail space performance measure or at least one space cost curve without user input. The use of automation where the system and method can pre-determine the best retail space performance measure or at least one space cost curve without user input reduce the effort and time to a user to produce optimal results for product allocation.

[0110] At steps 232 and 234, the performance curve module 150 determines the ‘best fit’ curve, for yield curves and cost curves respectively, based on an automated process, for every cluster product pair, given the magnitude of the number of curves from which to choose by considering various statistical tests. The system and method can pre-determine the ‘best-fit’ curves which reduce the effort and time to a user especially where the number of possible selections may be overwhelming or where there is not sufficient time to do iterative analysis to optimise results for determining the best fit manually.

[0111] The two sub-methods 200a and 200b conclude at step 236 where the product-space performance module 140 provides a performance curve for a cluster product combination comprising a yield curve combined with various cost curves.

[0112] At step 238, the product-space performance module 140 adjusts the performance curves from a cluster level to a retail store level resulting in store product level performance estimation for a given product or product category. Preferably, this adjustment of the performance curves is based on historical data or processed data and configuration parameters as selected by the user. By using the product-space performance curve module 140, each product considered for allocation of retail space in the retail store now will have such a set of retail space allocation values and their corresponding space performance measures (KPIs). In other words, the set of retail space allocation values relates to identification of the possible spaces a product can take within the store. For a fixed store size, this will lead to a significantlySUBSTITUTE SHEET (RULE 26)large solution space that cannot be computed in linear time. An optimisation process, as described in method 300 illustrated in Fig. 3, reduces this large solution space automatically by reduction of the number of data points to a more manageable number to produce more meaningful results to the user.

[0113] The optimisation method 300 has three sub-methods 300a, 300b and 300c. In the first sub-method 300a, in step 304, the set of retail space allocation values for each product and their associated at least one estimated performance measure is received by the optimisation engine module 160 as an input from the output determined at step 238 of method 200. At step 306, the optimisation engine module 160 is then adapted to determine the minimum and maximum retail allocation space values. The minimum and maximum space values are then used to establish the maximum and minimum bounds of the possible retail space allocation values. The possible retail space allocation values can also be bounded by the jumps from the minimum and maximum, and the possible space exceptions within those jumps. The module 160 is also adapted to arrange the retail space allocation values in sequence from minimum to maximum or to alternatively to take disjoint values in a range of retail space allocation values. At step 308, the optimisation engine module 160 is then adapted to identify any retail space exceptions and additions and to modify the set of retail space allocation values accordingly which results in the optimisation engine module 160 determining an initial set of possible retail allocation spaces at step 310. The jumps and exceptions are generally determined by possible planograming options available for the client teams to implement.

[0114] In the sub-method 300b, at step 312, the optimisation engine module 160 is adapted to begin a prioritisation process described as follows. At step 314, the optimisation engine module 160 is then adapted to determine at least one curve for the selected at least one performance measure which is based on the current retail space volume. At step 316, the optimisation engine module 160 is then adapted to identify valid minimum and maximum ranges and possible retail space values for exemption from the at least one curve determined at step 314. In one example, the optimisation engine module 160 uses historical data to fit the at least one curve and identify an optimal minimum and / or maximum within which the curves are valid from the valid minimum and maximum ranges identified. At step 318, the optimisation engine module 160 is then adapted to determine configurations at a store product level.SUBSTITUTE SHEET (RULE 26)

[0115] At step 320, the optimisation engine module 160 is then adapted to take the initial set of possible retail allocation spaces determined at step 310 and the configurations at a store product level at step 318 and to perform an LP domination check at step 320. This step 320 reduces number of data points that might never be needed to be traversed to achieve optimisation. When the dominance relationship is checked on continuous data, Linear Programming (LP) is the preferred mode. Hence the terminology LP un-dominated. The reduced final possible retail allocation space is determined and output at step 322. In the current situation, LP un-domination translates to no negative values in the curve and diminishing function throughout the range of the curve.

[0116] In the sub-method 300c, at step 324 the optimisation engine module 160 is adapted to receive the retail space which is to be optimised, i.e. a space of specific retail store. At steps 326, the optimisation engine module 160 is adapted to receive the product optimisation scope as selected by the user and then at step 328, is adapted to roll up to get store level space to optimise the selected performance space measure. In this example, the product optimisation scope refers to identification of all products, within a store, that are set to be optimised and roll up their space values so as to determine the total space that is available to optimise. This may also include space of products that are to be removed from the store. At step 330, the store level space is optimised and fixed. Similarly, to the optimisation of the selected performance space measure, the products that are to be fixed are identified so as to calculate the space taken by those products that is to be fixed in the store.

[0117] At step 322, the outputs of the three sub-methods 300a, 300b and 300c are all received by the optimisation engine module 160, namely, the final possible space set at step 322, the configurations at a store product level at step 318 and the store level space to optimise at step 330. The optimisation engine module 160 is then adapted to determine, at step 332, the selected at least one performance measure (KPI) with possible retail space allocation values and incremental at least one performance measure (KPI) per space lag and the retail space for optimisation. Retail allocation space for a product has valid maximum and minimum values within the set of retail allocation spaces as possible options. The difference between a space option and its previous option is defined as space lag.

[0118] At step 334, the optimisation engine module 160 is adapted to determine if the retail store space has a total minimum space allocation which is greater than the allocation space toSUBSTITUTE SHEET (RULE 26)be optimised. If yes, then the method moves to step 336 and the optimisation engine module 160 is adapted to only allocate the minimum space. However, if no, then the method moves to step 338 where the allocation of space is based on the selected at least one performance measure and is optimised using an optimisation function. Preferably, the optimisation function is an algorithm that uses the principles of multi choice knapsack greedy algorithm as discussed in more detail in the above paragraphs.

[0119] There is a possibility of fringe space left over at the end of the optimisation since an optimal solution is not always guaranteed due to conflicting constraints thus at step 342, the optimisation engine module 160 is adapted to determine if there is fringe space is left over. If yes, then at step 344, the optimisation engine module 160 is adapted to allocate fringe space, preferably using an allocation algorithm, for example the 0 1 knapsack algorithm which is discussed in more detail in the above paragraphs. Fringe space can be added where applicable after step 344 which results in optimised retail space allocation at the store product level. Finally, at step 348, the optimisation engine module 160 is adapted to output a final allocation of space for a list of products for a new or existing store for a maximised selected at least one performance measure. Preferably, the final allocation of space allocated to the products / product categories is in the form of an ordered steps of allocation of space. The optimisation engine module 160 is also adapted to determine the optimal size of a retail store or portion thereof, such as a department, zone, floor or aisle, by combining the final allocation of retail space for the list of products with store level per unit of space cost estimates. This ordered step data is the key ingredient to deliver on the fly scenarios and customisations that include future growth forecasts and other external change factors. Given the ordered nature of the data, it further helps move from an existing store planogram to the next most profitable planogram change. Such a change helps store planners decide the right order of moving from an existing planogram to the optimal planogram.

[0120] The preferred embodiments of the present invention provide repeatable and efficient determination of target product category product allocation to a designated space for the tens or hundreds of thousands of such recommendations required to optimise the allocations of product category space for a single store or multiple stores of a typical large multi-category retailer. Moreover, the embodiments of the invention provide a system and method which greatly automates the process of efficient and optimal product allocation to the designatedSUBSTITUTE SHEET (RULE 26)space such as a retail store with ease of use, reduced effort and time efficiency for the user. In addition, for a more advanced user, the system and method can provide customisation of all the parameters to assist adaptation of the optimisation process as necessary.

[0121] Upon implementation of the recommended target product / product category space allocations, benefits for a retail business include:Increases to sales and profit through increased customer purchases.Improved customer satisfaction linked to provision of a mix of product category space allocations that is more aligned to their needs and / or preferences in each store.Improved product availability which in turn further supports sales and profit growth through enhanced customer conversion.Reduced manual handling of products through the in-store supply chain to fixture and reduced need for effort to mark-down and clear slower moving inventory.More efficient allocation of working capital and higher rates of inventory / stock turns.

[0122] Furthermore, by being able to determine an optimal size of a retail store (the preselected space) for a particular geographic area, there are advantages in reducing wastage in building the right size of retail store, and associated land costs, which maximise total store performance thereby avoiding building a store, which are too big for the community or retailer needs, or not sufficiently large enough to service those needs.

[0123] In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word “comprise” or variations such as “comprises” or “comprising” is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.SUBSTITUTE SHEET (RULE 26)CLAIMS:1. A product allocation method (200) for performance by a computer having a processor and a memory, the method (200) comprising: storing, in a database (102, 110, 180) in communication with the computer, information in association with respective identifiers of spaces of a network of spaces of a product display and / or storage environment for displaying and / or storing a plurality of products; receiving, from a user device in communication with the computer over a computer network, a user input comprising a designated space, at least one performance measure and product data corresponding to a sub-set of products (steps 204 and 206, module 190); in response to receiving the product data, automatically retrieving information from the database and automatically determining a correspondence of the plurality of products for display and / or storage in the network of spaces with the sub-set of products in order to automatically identify at least one group of spaces from the network of spaces (step 210); in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure (steps 214 to 236); mapping, automatically, the determined assessment of the spaces in the at least one group to an assessment in the designated space (step 238, module 130); determining, automatically, a ranking of a plurality of product allocations of the designated space from the mapped assessment (subroutine 300, Fig. 3); determining, automatically, using the computer, allocation of products of the sub-set of products in the designated space of the product display and / or storage environment on the basis of the determined ranking (step 346).SUBSTITUTE SHEET (RULE 26)2. A method according to claim 1, wherein the step of automatically generating the space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure includes performing normalisation of values of the at least one performance measure for reduction of variation within the at least one group of spaces (step 212, module 125).3. A method according to claim 1 or claim 2, wherein the determined assessment of the spaces in the at least one group to an assessment in the designated space includes: calculating at least one function which fits the relationship between spaces of the at least one group of stores and the at least one performance measure (steps 222, 228, module 150); determining, automatically, a set of space allocation values allocated to the sub-set of products and their associated at least one performance measure from the at least one function (step 310, module 140); calculating a set of allocated space allocation values for each respective product of the subset (step 318, module 140).4. A method according to claim 3, wherein the relationship between spaces of the at least one group of stores and the at least one performance measure is calculated on the basis of historical information automatically retrieved from the database (step 224, module 150).5. A method according to any one of claims, wherein calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes one or both of the following: calculating a yield curve which determines volume sales performance in relation to variations in space allocation of the sub-set of products (steps 222, 228, 232, module 150); calculating a cost curve which determines performance measures in relation to the sub-set of products (steps 222, 228, 234, module 150).6. A method according to claim 5, wherein calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes calculating a plurality of a combinations of at least one yieldSUBSTITUTE SHEET (RULE 26)curve and at least one cost curve for each group of spaces and the sub-set of products (step 236, module 140).7. A method according to claim 6, wherein the method is adapted to determine which combinations have a pre-selected value of the at least one performance measure for optimal allocation of the subset of products to the designated space (step 236, module 140).8. A method according to claim 7, the method includes the step of calculating a list of the sub-set of products with their associated allocated space values that optimise the at least one performance measure using the combinations having a pre-selected value of the at least one performance measure (subroutine 300, module 160).9. A method according to claim 8, the method includes the step of determining an Ooptimal size of a retail store or portion of said store by combining said plurality of steps for allocating products to a space with space cost estimates (subroutine 300, module 160).10. A product allocation system (100) for performance, the system comprising: a computer having a processor and a memory; a database (110, 180) in communication with the computer, the database (110, 180) being adapted to store information in association with respective identifiers of spaces of a network of spaces of a product display and / or storage environment for displaying and / or storing a plurality of products; wherein the processor is adapted to execute a plurality of modules stored in the memory, each module being able to execute a set of instructions, and wherein the modules comprise: a spatial performance assessment module (120) adapted to: in response to receiving the product data, automatically retrieving information from the database and automatically determining a correspondence of the plurality of products for display and / or storage in the network of spaces with the sub-set of products in order to automatically identify at least one group of spaces from the network of spaces (module 120); in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set ofSUBSTITUTE SHEET (RULE 26)products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure; in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure; a product-space performance module (140) adapted to: mapping, automatically, the determined assessment of the spaces in the at least one group to an assessment in the designated space (module 130); determine, automatically, a ranking of a plurality of product allocations of the designated space from the mapped assessment; a space optimisation module (160) adapted to: determine, automatically, using the computer, allocation of products of the sub-set of products in the designated space of the product display and / or storage environment on the basis of the determined ranking.11. A system according to claim 10, including a display module adapted to: in response to receiving, from a user device (80) in communication with the computer over the computer network, a user input, determining a 2-D geographic map of how each product of the sub-set of products is allocated to the designated space; displaying, on the user device, the 2-D geographic map or a tabulation of how each product of the sub-set of products is allocated to the designated space.12. A system, method or a computer readable storage medium as described above wherein the designated space is a retail store of a physical retail environment, virtual retail environment or a space of a distribution centre.13. A method according to claim 1, including the step of displaying a 2-D geographic map of how each product of the sub-set of products is allocated to the designated space, upon receiving a user input.14. A method according to claim 1, including the step of determining a tabulation of how each product of the sub-set of products is allocated to the designated space.SUBSTITUTE SHEET (RULE 26)

Claims

34CLAIMS:

1. A product allocation method (200) for performance by a computer having a processor and a memory, the method (200) comprising: storing, in a database (102, 110, 180) in communication with the computer, information in association with respective identifiers of spaces of a network of spaces of a product display and / or storage environment for displaying and / or storing a plurality of products; receiving, from a user device in communication with the computer over a computer network, a user input comprising a designated space, at least one performance measure and product data corresponding to a sub-set of products (steps 204 and 206, module 190); in response to receiving the product data, automatically retrieving information from the database and automatically determining a correspondence of the plurality of products for display and / or storage in the network of spaces with the sub-set of products in order to automatically identify at least one group of spaces from the network of spaces (step 210); in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure (steps 214 to 236); mapping, automatically, the determined assessment of the spaces in the at least one group to an assessment in the designated space (step 238, module 130); determining, automatically, a ranking of a plurality of product allocations of the designated space from the mapped assessment (subroutine 300, Fig. 3); determining, automatically, using the computer, allocation of products of the sub-set of products in the designated space of the product display and / or storage environment on the basis of the determined ranking (step 346).

352. A method according to claim 1, wherein the step of automatically generating the space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure includes performing normalisation of values of the at least one performance measure for reduction of variation within the at least one group of spaces (step 212, module 125).

3. A method according to claim 1 or claim 2, wherein the determined assessment of the spaces in the at least one group to an assessment in the designated space includes: calculating at least one function which fits the relationship between spaces of the at least one group of stores and the at least one performance measure (steps 222, 228, module 150); determining, automatically, a set of space allocation values allocated to the sub-set of products and their associated at least one performance measure from the at least one function (step 310, module 140); calculating a set of allocated space allocation values for each respective product of the subset (step 318, module 140).

4. A method according to claim 3, wherein the relationship between spaces of the at least one group of stores and the at least one performance measure is calculated on the basis of historical information automatically retrieved from the database (step 224, module 150).

5. A method according to any one of claims, wherein calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes one or both of the following: calculating a yield curve which determines volume sales performance in relation to variations in space allocation of the sub-set of products (steps 222, 228, 232, module 150); calculating a cost curve which determines performance measures in relation to the sub-set of products (steps 222, 228, 234, module 150).

6. A method according to claim 5, wherein calculating at least one function which fits the relationship between allocation of space for the subset of products and the at least one performance measure includes calculating a plurality of a combinations of at least one yieldcurve and at least one cost curve for each group of spaces and the sub-set of products (step 236, module 140).

7. A method according to claim 6, wherein the method is adapted to determine which combinations have a pre-selected value of the at least one performance measure for optimal allocation of the subset of products to the designated space (step 236, module 140).

8. A method according to claim 7, the method includes the step of calculating a list of the sub-set of products with their associated allocated space values that optimise the at least one performance measure using the combinations having a pre-selected value of the at least one performance measure (subroutine 300, module 160).

9. A method according to claim 8, the method includes the step of determining an Ooptimal size of a retail store or portion of said store by combining said plurality of steps for allocating products to a space with space cost estimates (subroutine 300, module 160).

10. A product allocation system (100) for performance, the system comprising: a computer having a processor and a memory; a database (110, 180) in communication with the computer, the database (110, 180) being adapted to store information in association with respective identifiers of spaces of a network of spaces of a product display and / or storage environment for displaying and / or storing a plurality of products; wherein the processor is adapted to execute a plurality of modules stored in the memory, each module being able to execute a set of instructions, and wherein the modules comprise: a spatial performance assessment module (120) adapted to: in response to receiving the product data, automatically retrieving information from the database and automatically determining a correspondence of the plurality of products for display and / or storage in the network of spaces with the sub-set of products in order to automatically identify at least one group of spaces from the network of spaces (module 120); in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set ofproducts in each space of the identified at least one group of spaces on the basis of the at least one space performance measure; in response to identifying the at least one group of spaces, automatically generate a space performance assessment of allocation of each product of the sub-set of products in each space of the identified at least one group of spaces on the basis of the at least one space performance measure; a product-space performance module (140) adapted to: mapping, automatically, the determined assessment of the spaces in the at least one group to an assessment in the designated space (module 130); determine, automatically, a ranking of a plurality of product allocations of the designated space from the mapped assessment; a space optimisation module (160) adapted to: determine, automatically, using the computer, allocation of products of the sub-set of products in the designated space of the product display and / or storage environment on the basis of the determined ranking.

11. A system according to claim 10, including a display module adapted to: in response to receiving, from a user device (80) in communication with the computer over the computer network, a user input, determining a 2-D geographic map of how each product of the sub-set of products is allocated to the designated space; displaying, on the user device, the 2-D geographic map or a tabulation of how each product of the sub-set of products is allocated to the designated space.

12. A system, method or a computer readable storage medium as described above wherein the designated space is a retail store of a physical retail environment, virtual retail environment or a space of a distribution centre.

13. A method according to claim 1, including the step of displaying a 2-D geographic map of how each product of the sub-set of products is allocated to the designated space, upon receiving a user input.

14. A method according to claim 1, including the step of determining a tabulation of how each product of the sub-set of products is allocated to the designated space.

Images