WO2015059807A1 - 情報処理システムおよび情報処理方法 - Google Patents
情報処理システムおよび情報処理方法 Download PDFInfo
- Publication number
- WO2015059807A1 WO2015059807A1 PCT/JP2013/078894 JP2013078894W WO2015059807A1 WO 2015059807 A1 WO2015059807 A1 WO 2015059807A1 JP 2013078894 W JP2013078894 W JP 2013078894W WO 2015059807 A1 WO2015059807 A1 WO 2015059807A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- customer
- information
- shelf
- store
- probability
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Definitions
- the present invention relates to an information processing system and an information processing method. More specifically, the present invention relates to an information processing system and an information processing method for simulating customer trends in a store.
- Patent Documents 1 and 2 As background art in this technical field.
- Patent Document 1 calculates the residence time for each place and the probability of the next place to move based on the movement information of a car or a person having a position terminal (such as GPS). Including, a technique for estimating a flow line is described.
- Patent Document 2 describes a place that is easy to enter the human field of view by walking an agent (entering the breadth, height, and movement path of the field of view) on the 3D map information. A technique is described in which the degree of the place (natural monitorability) is estimated and thereby a crime prevention simulation is performed.
- the product arrangement in a store is determined depending on the intuition and experience of the person in charge at the site. Because the factors that determine purchase are complex, such as product characteristics, customer behavior characteristics (stay time, number of points purchased, etc.), location characteristics (shelf position and height), etc. This is because they are considered to understand these factors most. That is, conventionally, the possibility of purchasing the product M at the coordinate P has a relationship represented by the following formula (1) with f as a function, and the person in charge understands this most. Based on the idea, the product placement of the store was decided.
- an object of the present invention is to provide a technique that makes it easier to determine the product arrangement.
- the present application includes a plurality of means for solving the above-described problems, and an example thereof is as follows.
- the first information regarding the stay probability with respect to the passage of time after the customer enters the store the second information indicating the distance between each of the plurality of shelves provided in the store, and ,
- the customer is D)
- the customer stays in the store only for the dwell time
- the customer stores the simulation condition that the destination shelf is moved randomly, the first information, the second information, the third information, and Using multiple simulation conditions Wherein the simulator to calculate the probability that the customer will stay a display unit for displaying in association probabilities to shelves, to have a relative respectively.
- the information processing method Comprising: 1st information regarding the stay probability with respect to the time passage after a customer enters a store, 2nd information which shows the distance between each of the some shelf provided in the store, and A step of receiving, as input, a staying time at which the customer stays in the store and a third information indicating a moving period in which the customer moves between shelves, the first information, the second information, and the third information; And a) the customer starts moving from the entrance of the store, b) the customer is more likely to move to a shelf that is farther away from the shelf, c) the customer D) calculating the probability of the customer staying with respect to each of the plurality of shelves using the simulation condition that the customer is randomly moving the destination shelf; Process to display in association with the shelf , Characterized by having a.
- a store unit for storing linked information, shelf coordinate information, shelf number of the shelf, and information associated with these are used to determine a stay position or a stay probability of a store customer at a certain time t.
- the second process for obtaining the stay position or stay probability at the customer's time (t + ⁇ t) are repeated a plurality of times, thereby calculating the customer's ease of visit for each shelf or the sales forecast for each shelf.
- the person in charge of product placement can more easily determine the product placement.
- factors that determine the possibility of purchasing the product M at the coordinate P include the location characteristics, the product characteristics, and the customer behavior characteristics.
- the inventors of the present application paid particular attention to the location characteristics among the above factors.
- the customer simulator system includes store layout evaluation content and product shelf layout optimization content.
- the customer simulator uses the simulation results of the customer simulator.
- the store layout evaluation content is a content that predicts the customer flow line that excludes the influence of the product and the ease of stopping by, for example, predicting the customer stop for the layout plan when opening a new store or changing the store layout. Is possible.
- the merchandise shelf arrangement optimization content makes it possible to predict, for example, the increase / decrease in the customer unit price, the number of purchase points, and the number of purchased items due to the change in the shelf arrangement, in addition to the customer stoppage.
- FIG. 1 shows a system outline of the first embodiment.
- the user (US) can browse the content (K) by operating the client (CL).
- the client (CL) is connected to the network (NW), and transmits a request from the user (US) to the application server (AS) via the client (CL).
- the application server (AS) performs processing based on the request of the user (US) and transmits the result to the client (CL).
- the client (CL) generates a screen using the received result and displays it on the content (K) of the display (CLID).
- FIG. 2A and FIG. 2B are explanatory diagrams showing components of a customer simulator system according to one embodiment, and are divided for convenience of illustration, but the respective processes shown in FIG. Executed.
- FIGS. 2A and 2B show a series of flows from an application server (AS) that performs processing of the customer simulator system to a client (CL) that outputs an analysis result to a viewer.
- AS application server
- CL client
- This system includes an application server (AS) and a client (CL). Each of them has a normal computer configuration including a processing unit, a storage unit, a network interface, and the like.
- the application server (AS) shown in FIG. 2A executes a customer simulator process.
- the application server (AS) receives a request from the client (CL) shown in FIG. 2B, the application is started automatically or manually at the set time.
- the result analyzed by the application server (AS) is transmitted through the network (NW) to the client (CL) shown in FIG. 2B.
- the application server includes a transmission / reception unit (ASS), a storage unit (ASM), and a control unit (ASC).
- ASS transmission / reception unit
- ASM storage unit
- ASC control unit
- the transmission / reception unit transmits and receives data to and from the client (CL) shown in FIG. 2B. Specifically, the transmission / reception unit (ASS) receives a command sent from the client (CL), performs a customer excursion simulation at the control unit (ASC), and transmits an analysis result to the client (CL).
- the storage unit (ASM) is composed of an external recording device such as a hard disk, memory or SD card.
- the storage unit (ASM) stores a database for executing simulation, setting conditions, and results.
- the storage unit (ASM) stores a simulation database (D), a sales database (E), a shelf database (F), a map database (G), a stop-by database (H), and an accounting database (I). .
- the simulation database (D) is a database that stores parameters and output results necessary for executing the simulation.
- the sales database (E) is a database that stores data related to purchase such as POS data.
- the shelf database (F) is a database that stores data related to the shelf.
- the map database (G) is a database that stores data relating to maps such as shelf arrangement.
- the drop-in database (H) is a database that stores data related to customer's products and shelves.
- the charging database (I) is a database that stores data related to charging when the user (US) uses the customer simulator.
- the control unit includes a central processing unit CPU (not shown), and executes control of data transmission / reception and simulation. Specifically, a CPU (not shown) executes a program registered in advance in the control unit (ASC).
- Communication control controls the timing of communication with a client (CL) by wire or wireless. Further, the communication control (ASCC) executes data format conversion and sorting of destinations by data type.
- the customer simulator (AP) is a process for selecting necessary data from data registered in the storage unit (ASM) based on a request from the client (CL) and performing a simulation.
- the customer simulator (AP) includes store layout evaluation (APA), drop-in simulation (APB), store layout evaluation learning (APC), merchandise shelf replacement calculation (APD), and billing (APE).
- Store layout evaluation is a process of evaluating a layout by dividing the set shelf arrangement and product into a place effect and a product effect.
- the stop-by simulation is a process for obtaining the stop-off rate from the simulation from the set shelf arrangement.
- Store layout evaluation learning is a process of learning the rack placement and stop-by rate in actual measurement, and parameters related to the drop-in in that industry.
- the product shelf replacement calculation is a process for predicting sales by selecting a shelf and a product using the store layout evaluation (APA).
- Charge (APE) is a process of charging when the user (US) uses the customer simulator.
- the Web server (ASCW) performs processing for controlling access IO from the client (CL).
- the client (CL) passes through the Web server (ASCW) when obtaining the setting information. Further, the result of the customer simulator (AP) is transmitted to the client (CL) via the Web server (ASCW).
- the analysis result is stored in the simulation database (D) and transmitted to the client (CL) shown in FIG. 2B through the transmission / reception unit (ASSR).
- ASSR transmission / reception unit
- the client (CL) shown in FIG. 2B is a contact point with the user, and inputs and outputs data.
- the client (CL) includes an input / output unit (CLI), a transmission / reception unit (CLS), a storage unit (CLM), and a control unit (CLC).
- the input / output unit (CLI) is a part serving as an interface with the user.
- the input / output unit (CLI) includes a display (CLID), a keyboard (CLIK), a mouse (CLIM), and the like. Other input / output devices can be connected to an external input / output (CLIU) as required.
- the display is an image display device such as a CRT (CATHODE-RAY TUBE) or a liquid crystal display.
- the display (CLID) may include a printer or the like.
- the transmission / reception unit (CLS) transmits and receives data to and from the application server (AS) shown in FIG. 2A. Specifically, the transmission / reception unit (CLS) transmits the analysis condition information (CLMP) to the application server (AS) and receives the analysis result.
- CLMP analysis condition information
- the storage unit (CLM) is composed of an external recording device such as a hard disk, memory or SD card.
- the storage unit (CLM) records information necessary for analysis and drawing, such as analysis condition information (CLMP) and drawing setting information (CLMT).
- CLMP analysis condition information
- CLMT drawing setting information
- Analysis condition information records conditions such as the number of members to be analyzed set by the user and analysis method selection.
- the drawing setting information (CLMT) records information on the drawing position such as what is plotted in which part of the drawing. Furthermore, the storage unit (CLM) may store a program executed by a CPU (not shown) of the control unit (CLC).
- the control unit includes a CPU (not shown) and executes communication control, input of analysis conditions from the client user (US), drawing for presenting the analysis result to the client user (US), and the like. To do. Specifically, the CPU executes a program stored in the storage unit (CLM), thereby performing communication control (CLCC), Web browser (CLCW), analysis setting (CLCT), drawing setting (CLCP), and content generation. (CLCA) processing is executed.
- CLCC communication control
- CLCW Web browser
- CLCT analysis setting
- CLCP drawing setting
- CLCA content generation.
- Communication control controls the timing of communication with the application server (AS) by wire or wireless.
- the communication control converts the data format and distributes the destination according to the data type.
- the Web browser (CLCW) is an interface with the user (US), sets analysis condition information (CLMP) and drawing setting information (CLMT), and generates content (CLCA) from the result of the application server (AS). ) Is displayed on the Web browser (CLCW).
- the analysis condition (CLCT) receives an analysis condition designated from the user via the input / output unit (CLI) and records it in the analysis condition information (CLMP) of the storage unit (CLM).
- CLMP analysis condition information
- the client sends these settings to the application server (AS), requests analysis, and executes drawing settings (CLCP) in parallel therewith.
- the drawing setting (CLCP) calculates a method for displaying the analysis result based on the drawing setting information (CLMT) and a position for plotting the drawing.
- the result of this processing is recorded in the drawing setting information (CLMT) of the storage unit (CLM).
- CLCA Content generation
- AS application server
- CLMT drawing setting information
- CLOD display
- CLCW Web browser
- FIG. 3 is a flowchart of a stop-by simulation (APB) of the customer simulator (AP).
- This process is a process for obtaining a stop-by rate S (probability of a customer staying on each of a plurality of shelves), which is an element for obtaining a location bias that is a property of a location.
- This process is a partial flow of the store layout evaluation (APA) described with reference to FIG. 4.
- APA store layout evaluation
- API store layout evaluation learning
- the stopover rate is obtained by simulation under the hypothesis that a shelf with a high stopover rate S is a shelf in a place where it is easy to stop by.
- API1 When the process is started (APB1), the necessary input file is read by input (APB2).
- APIB2 Necessary input files are information on location (shelf arrangement, entrance / exit location, etc.) and information on customer behavior characteristics.
- the information regarding the location is information indicating the distance between each of the plurality of shelves provided in the store, and is usually known from the store layout information and the like.
- information on customer behavior characteristics is information obtained by measuring a customer's movement route (information that associates time and shelf position) with various sensors such as a video camera and a wearable sensor. It is the information regarding the stay probability with respect to the passage of time after entering the store.
- the inventors of the present application conducted a demonstration experiment on the behavioral characteristics of customers in a store, and a customer in front of one shelf moves to another shelf (hereinafter referred to as “hopping”). ) The probability is lower as the distance between the shelves (the effective distance considering the distance of the detour when there are obstacles between the shelves) is lower.
- the behavioral characteristics of this customer follow the so-called “exponential distribution” where the horizontal axis is the rank of the movement distance between shelves (the customer who moved farther is the right side of the graph) and the vertical axis is the number of people who moved the distance.
- the vertical axis of this graph is a logarithmic axis, the behavior can be approximated by a straight line.
- the slope and intercept of the exponential distribution are known, it can be understood that the customer behavior characteristics are uniquely determined. Therefore, it is possible to make the customer's behavior constant by the binary value of the distribution slope and intercept.
- This intercept determines the customer's stay time offset (the time that most of the customers stay in the store uniformly), and the slope determines the stay time (the stay probability is 1 / e for each time). In this way, the customer behavior is made constant by the stay time offset and the stay time.
- the characteristics of the product may affect the behavior of the customer, such as the frequency of visiting a certain place being abnormally high depending on the particular popular product, but the above knowledge is only probabilistic of the customer's behavior Because it focuses only on the special characteristics, the influence of the product is eliminated.
- the stay time offset (DP5) for example, the stay time (DP6), the movement period (in the parameter table (DP) of the corresponding project ID (DP1) for reading the input file necessary for the stop-by simulation (APB) ( DP7), travel distance (DP8), simulation time (DP9), and shelf distance table (FD) of shelf database (F) are read. Details of these tables will be described later (details of each table are the same below).
- Step 1 A state transition is obtained.
- tr (i, j) exp ( ⁇ dd (i, j) / beta)
- tr (i, j) is the state transition probability
- dd (i, j) is the inter-shelf distance table (FD)
- beta is the movement distance (DP8).
- the state transition probability matrix (APB4) is a result output by the state transition probability (APB3). The contents will be described with reference to FIG.
- Probability by hopping is the probability of going to the shelf for each shelf for each hopping.
- the calculation formula is composed of four steps. (Step 1.) Determine initial conditions.
- weight is added to the entrance. This expresses the behavior that the customer is at the entrance at the start of the simulation. Specifically, a weight is added to the entrance and the initial stop shelf by the icon type (GM7) of the map table (GM) of the map database (G). This is expressed as follows.
- pm (j, 1) 1
- pm (j, 1) is a probability table for each hopping.
- 1 is the first hopping frequency, that is, start. Even if the store has a plurality of entrances, this influence is absorbed by normalization, so the weight of each of the plurality of entrances may be 1.
- Step 2. The probability of going to the shelf j when the number of hoppings is k times is determined.
- pm (j, k) pm (i, k-1) * tr (i, j)
- tr (i, j) is a state transition probability
- pm (j, k) is a probability table for each hopping.
- the formula for the probability by hopping (APB5) is an example, and other calculation formulas may be used.
- the hopping probability array (APB6) is the result output by the hopping probability (APB5). Details will be described with reference to FIG.
- the probability of going to the shelf j in the cumulative hopping is obtained.
- the calculation formula is composed of four steps. (Step 1.) 0 is substituted as an initial condition. (Step 2.) The probability that the number of hops will go to the shelf j by k times is obtained by the following equation.
- cc (j, k) cc (j, k-1) + (1-cc (j, k-1)) * pm (j, k)
- cc (i, k) is a cumulative probability. That is, (probability that the number of hopping has stopped at shelf j by k-1 times) + (probability that hopping number has not stopped at shelf j by k-1 times) * (stops at shelf j when hopping number is k times) Probability).
- Step 3. Repeat 2 until the total number of hops.
- the value when cc is the total number of hoppings is output as the stop-by rate S (APB8).
- the stopover rate S (APB8) is a result output by the cumulative hopping probability (APB7). Details will be described with reference to FIG.
- a network diagram between products may be created from the correlation of each product of daily sales from the POS table (EP), and the coefficient obtained from this may be included in the drop-in simulation (APB).
- a node on the network indicates a product, and an edge indicates a relationship. By combining these, it becomes a model in which the frequency of movement differs depending on whether or not they are close to each other on the network.
- FIG. 4 is a flowchart of the store layout evaluation (APA) of the customer simulator (AP).
- an input file necessary for store layout evaluation (APA2) is read.
- the input file is data corresponding to a desired item ID (DP1, FD1) in the parameter table (DP) and the inter-shelf distance table (FD).
- APA3 In the drop-in simulation (APA3), the simulation described with reference to FIG. 3 is performed using the data of the input (APA2).
- the stopover rate S (APA4) is an output result of the stop-by simulation (APA3).
- the location bias processing obtains the location effect using the stopover rate S (APA4) and the stop-by model (APA6) obtained by the store layout evaluation learning (APC) described in FIG.
- the stop model (APA6) is the same as the stop model (DP10) of the parameter table (DP).
- location bias 1 / (1 + exp (-1 * (stop rate S * slope + intercept)))
- Location bias (APA7) is an output result of location bias processing (APA5). The contents are shown in FIG.
- APA8 the effect of the product group excluding the place effect from the sales (APA9) (hereinafter, “product effect”) is obtained.
- the inputs of the bias processing (APA8) are the location bias (APA7) and the sales (APA9).
- Sales (APA9) is equivalent to the sales table (EU) of the sales database (E).
- the product effect (APA10) is an output result of the bias processing (APA8), and shows the effect of the product group excluding the place effect.
- FIG. 5 is a sequence diagram of the store layout evaluation content using the customer simulator.
- FIG. 5 includes a client (CL), an application server (AP), and an application server (AP) recording unit (APM).
- CL client
- AP application server
- API application server
- APM application server recording unit
- Each vertical arrow means the order of processing in time series.
- each horizontal line arrow indicates the relationship between each component.
- the server of the application server (AP) is activated to make the access acceptable from the client (CL).
- Application activation (CL1) indicates that the user (US) has activated the store layout evaluation content.
- the condition input (CL2) sets conditions until the customer simulator is executed. It is executed under the analysis condition (CLCT) of the client (CL) and recorded in the analysis condition information (CLMP).
- CLCT analysis condition
- CLMP analysis condition information
- the application server (AP) is instructed to start the store layout evaluation content.
- the charging database (I) side updates the charging table (IK). If charging is performed based on the number of clicks (IK4), the corresponding item in the charging table (IK) is incremented by one. If charging is performed based on the cloud usage time (IK5), the start time is recorded for the corresponding item in the charging table (IK). This is the update (I1). Furthermore, in the case of termination, the number of clicks (IK4) is not counted, or the usage time is obtained from the start time and end time of the cloud usage time (IK5), and the value is used as the usage time. Add to IK5).
- condition transmission from the application server (AP)
- parameter table (DP) of the simulation database (D) is referred to from the analysis condition information (CLMP)
- ASM analysis condition information
- DFE1 conditional search
- the store layout evaluation (APA) the store layout evaluation (APA) shown in FIG. 4 is performed.
- content generation (CLCA) the result of the store layout evaluation (APA) is created using the drawing setting (CLCP) transmitted to the client (CL).
- CLCP drawing setting
- the end (CL5) is the end of the store layout evaluation content.
- the customer simulator system is configured such that the first information (customer behavior characteristic) regarding the stay probability with respect to the passage of time after the customer enters the store, each of the plurality of shelves provided in the store Second information (information about the location) indicating the distance between the two, and third information indicating the staying time (DP6) in which the customer stays in the store and the moving period (DP7) in which the customer moves between shelves And the input unit (transmission / reception unit ASS), the first information, the second information, and the third information, and a) the customer starts moving from the entrance of the store, b) the customer , Of the plurality of shelves, the probability of moving to a shelf closer to the farther away is high, c) the customer stays in the store only for the staying time, d) the customer randomly moves the destination shelf Memorize the simulation conditions And a simulator unit (customer simulator) that calculates the probability of a customer staying on each of a plurality of shelves using the storage unit, the first information, the second information, the third information,
- the information processing method Comprising: 1st information regarding the stay probability with respect to the time passage after a customer enters a store, 2nd information which shows the distance between each of the some shelf provided in the store, and A step of accepting as input, the first information, the second information, and the third information indicating the stay time in which the customer stays in the store and the movement cycle in which the customer moves between shelves, The third information, and a) the customer starts moving from the entrance of the store, b) the customer is more likely to move to a shelf that is farther away from the shelf, c) Using the simulation condition that the customer stays in the store only for the stay time, and d) the customer randomly moves the destination shelf, the probability that the customer stays for each of the plurality of shelves is calculated.
- Process Calculation execution CL3 , Characterized by having a step (content generation CLCA) for displaying in association probabilities to the shelf, the.
- the information processing system and information processing method according to the present embodiment can separate sales into place bias, which is a place characteristic, and product effect, which is a product characteristic.
- place bias which is a place characteristic
- product effect which is a product characteristic.
- FIG. 6 is a screen of the store layout evaluation content (KA).
- KA1 is a store name (DP2).
- KA2 is a calculation execution start button of the store layout evaluation simulation. This is the same as the calculation execution (CL3) in the sequence diagram of FIG.
- KA3 is a parameter in the parameter table (DP) of the simulation database (D).
- KA4 is a graph that displays a scatter diagram of location bias and product effects.
- KA5 is a setting screen for editing the shelf layout.
- KA6 is a shelf layout.
- KA7 is a screen that displays a stop-by rate that is a result of the store layout evaluation simulation.
- KA8 is a legend for shelf layout.
- KA9 is a screen for displaying the unit price of the customer as a result of the shelf layout evaluation simulation.
- KA10 is a screen that displays the number of customer purchase points as a result of the shelf layout evaluation simulation.
- KA11 is a screen for displaying the number of customer purchase
- FIG. 6 is a screen at the time of start-up. From this screen, the layout of the shelf is determined using KA5 and KA6. 7 indicates the result of calculation execution (CL3).
- the result of the location bias which means the location characteristics
- the result of the location bias is displayed.
- the magnitude of the location bias is represented by the color density arranged on the shelf in the shelf layout, and the dark color has a larger location bias value. That is, KB61 ⁇ KB62 ⁇ KB63. Each value is indicated in the legend KB8.
- FIG. 7 with the configuration in which only the location bias is associated with each of the plurality of shelves, the influence of the location effect on the store layout can be intuitively understood by the person in charge of determining the product placement. It becomes easier.
- FIG. 8 shows the screen when the replacement candidate for the stop of KB5 is selected with respect to the screen of FIG.
- the product effect that is the product characteristic is further superimposed.
- KC in FIG. 8 shows shelf replacement candidates and characteristics when shelves are replaced.
- KC4 is a scatter diagram of location bias and product effect, and each node KC41 to 43 means a product arranged on the shelf.
- KC41 is a shelf that has a higher product effect than average and a lower location bias than average.
- KC42 is a shelf with a product effect lower than average and location bias higher than average.
- KC43 is a shelf other than the above.
- KC6 is the evaluation result in the shelf layout. The meaning depends on the shelves. For each average of product effect and location bias, KC61 is a shelf with product effect higher than average and location bias lower than average. KC62 is a shelf with a product effect lower than average and location bias higher than average. KC63 is a shelf other than the above.
- the screen of FIG. 9 shows the result of moving (replacement) the arrangement of the products on the shelf of KC 6 of FIG.
- the replaced shelves are KD61 and KD62.
- KD9 the result of recalculating KD9, KD10, and KD11 is displayed. Furthermore, the difference from the initial value is also displayed.
- KD9 the result of calculating the customer unit price shown below from the sales model (APD4) created in FIG. 10 is displayed. This process is performed for all the shelves, and a total value is calculated by substituting the replaced location bias values for the shelves that have changed.
- the sales model (DP11) of the parameter table (DP) stores the slope and intercept of the customer unit price formula shown below.
- KD9 further displays not only the customer unit price, but also the amount of change and the rate of increase / decrease as the amount of change in the customer unit price before and after replacement.
- KD10 is the result obtained by replacing the customer unit price of KD9 with the number of customer purchase points.
- KD11 is the result obtained by replacing the customer unit price of KD9 with the number of customer purchase items.
- FIG. 10 is a flowchart of product shelf replacement calculation (APD) of the customer simulator (AP).
- APD merchandise shelf replacement calculation
- sales are predicted by processing the merchandise effect, which is the characteristics of the merchandise, and the location bias, which is the characteristics of the shelves (place), with the sales model.
- the location bias (APA7) and product effect (APA10) files are read.
- the sales model generation (APD3) regression based on the input (APD2) is performed. By generating the sales model, it is possible to predict the sales after the shelves are arranged.
- FIG. 11 is a flowchart of store layout evaluation learning (APC) of the customer simulator (AP). This is learning for obtaining a stop-by model (APA6) for performing store layout evaluation (APA).
- API store layout evaluation learning
- APA6 stop-by model
- an input file necessary for the stop-by simulation is read.
- the inter-shelf distance table (FD) of the shelf database (F) is read.
- the simulation is performed using the data of the input (APC2) as in FIG.
- a drop simulation APC3 is performed.
- the stop-off rate S APC4 is an output result of the stop-by simulation (APC3).
- APC5 regression based on the stop rate S (APC4) and the stop rate (APC6) is performed.
- APC4 regression based on the stop rate S
- APC6 stop rate
- the regression equation obtained in the stop model generation (APC5) is the stop model (APC7).
- This drop-in model (APC7) is substituted into the drop-off model (DP10), and the store layout evaluation learning (APC) is completed (APC8).
- FIG. 12 is a parameter table (DP) that stores parameters necessary for the customer simulator (AP).
- the case ID (DP1) is an ID for identifying the case.
- the project name (DP2) is a project name.
- Store No (DP3) is a number for identifying a store.
- the date and time (DP4) is the date on which the simulation is performed. If it extends over multiple days, multiple days may be specified. If date and time are required, both may be stored (the same applies to dates below).
- the stay time offset (DP5) is a value that serves as an offset when the simulation is executed. The unit of the value is seconds.
- the staying time (DP6) is the staying time of the customer when the simulation is executed, and is a value at which the staying probability of the customer becomes 1 / e for each time. The unit of the value is seconds.
- the moving cycle (DP7) is an average moving cycle from the previous shelf to the next shelf when the simulation is executed.
- the unit is seconds.
- the movement distance (DP8) is an average movement distance from the front shelf to the next shelf when the simulation is executed.
- the unit is meters.
- the simulation time (DP9) is a time for executing the simulation.
- the unit is seconds.
- the drop-in model (DP10) is a model parameter used in the location bias process.
- the model is constituted by values of various parameters of the fitting function or expressions of the fitting function itself.
- the sales model (DP11) is a model parameter used in the location bias process.
- the model is similarly configured with values and formulas.
- FIG. 13 is a state transition probability matrix table (DM) that stores state transition probabilities indicating the ease of stopping on the shelf.
- DM state transition probability matrix table
- the case ID (DM1) is an ID for identifying the case.
- the date and time (DM2) is the date on which the simulation is performed.
- the shelf ID 1 (DM 3) is a number for identifying the shelf 1
- the shelf ID 2 (DM 4) is a number for identifying the shelf 2.
- shelf ID1 (DM3) and shelf ID2 (DM4) are divided into frontage (stage (horizontal) and column (vertical)), they may be stored so that they can be identified.
- the state transition probability (DM5) stores the output result of the state transition probability (APB3) of the stop-by simulation (APB).
- FIG. 14 is a hop-by-hop probability table (DH) that stores the drop-in probability of each hop-by-hop shelf. Since this is also data used for simulation, it is included in the simulation database (D). Next, the contents will be described.
- DH hop-by-hop probability table
- the case ID (DH1) is an ID for identifying the case.
- the date and time (DH2) is the date on which the simulation is performed.
- the hop count (DH3) is the number of times hopping is repeated.
- the maximum value of the number of hoppings is a value obtained by dividing the simulation time (DP9) by the movement period (DP7).
- the shelf ID (DH4) is a number for identifying the shelf.
- the hop-by-hop probability (DH5) is a calculation result of the hop-by-hop probability (APB5) in the stop-by simulation (APB).
- FIG. 15 is a stop-by rate S table (DT) for storing the drop-in rate for each shelf as a result of the simulation.
- DT stop-by rate S table
- the case ID (DT1) is an ID for identifying the case.
- the date and time (DT2) is the date on which the simulation is performed.
- the shelf ID (DT3) is a number for identifying the shelf.
- the stop-off rate S (DT4) is a calculation result of a hop cumulative probability (APB6) described later in the stop-by simulation.
- FIG. 16 is a location bias table (DB) obtained by removing the influence of the product from the simulation result and obtaining only the location effect.
- DB location bias table
- the case ID (DB1) is an ID for identifying the case.
- the date and time (DB2) is the date on which the simulation is performed.
- the shelf ID (DB3) is a number for identifying the shelf.
- the location bias (DB4) is a calculation result of the location bias processing (APA5) of the store layout evaluation (APA).
- FIG. 17 is a product effect product effect table (DU) obtained by removing the influence of the place from the sales and obtaining the product effect.
- the case ID (DU1) is an ID for identifying the case.
- the date and time (DU2) is the date on which the simulation is performed.
- the shelf ID (DU3) is a number for identifying the shelf.
- the product ID (DU4) is a number for identifying the product.
- the product effect (DU5) is a calculation result of the bias processing (APA8) of the store layout evaluation (APA).
- FIG. 18 is a POS table (EP) for obtaining sales for each customer.
- the date and time (EP1) is the date and time when the product was registered at the cash register. That is, the date and time of purchase.
- the customer ID (EP2) is a number for identifying the purchased customer.
- the product ID (EP3) is a number for identifying the purchased product.
- Product information (EP4) is product information of product ID (EP3). It only needs to know the details of the product, and it doesn't have to be language information such as a barcode.
- the unit price (EP5) is a price per product ID (EP3).
- the number (EP6) is the number of products purchased with the product ID (EP3). Store No.
- EP7 is a number for identifying a store.
- the cash register No. (EP8) is a number for identifying the cash register in the store No. (EP7).
- the receipt No. (EP9) is a number for identifying the purchased product at the register No. (EP8) for each transaction.
- the account of one customer can be specified by combining the store number (EP7), the cash register number (EP8), and the receipt number (EP9).
- FIG. 19 is a sales table (EU) showing sales of products. This is a total of sales for each product from the POS table (EP).
- EU sales table
- a case ID (EU1) is an ID for identifying a case.
- the date and time (EU2) is the date on which the simulation is performed. That is, the date and time of purchase.
- the product ID (EU3) is a number for identifying the product.
- the product information (EU4) is product information of a product ID (EU3). It only needs to know the details of the product, and it doesn't have to be language information such as a barcode.
- the sales amount (EU5) is the sales amount for each product ID at the date and time (EU2), and the sales for each product at the target date and time (EU2) is represented by the unit price (EP5) and the number (EP6) of the POS table (EP). Tables stored in the shelf database (F) will be described with reference to FIGS.
- FIG. 20 is a shelf product table (FT) showing that a product and a shelf are related to each other. By using this, it is possible to know which product is placed on which shelf.
- the case ID (FT1) is an ID for identifying the case.
- the date and time (FT2) is the date on which the simulation is performed.
- the shelf ID (FT3) is a number for identifying the shelf. If the shelf is divided into frontage (stage (horizontal) and row (vertical)), it may be stored so that it can be identified (the same applies hereinafter).
- the number of products installed on the shelf (FT4) describes how many different products are arranged on the shelf. For example, when two types of products “ice” and “frozen food” are handled on the shelf “A”, the value is 2.
- the product ID (FT5) is a number for identifying the product.
- the same product installation shelf number (FT6) describes the number of shelves handled for the product ID when the same product is handled by a plurality of shelves. For example, when the product “Ice” is handled on the shelves “A”, “B”, and “C”, it becomes 3. By holding such values, the sales per shelf can be obtained by dividing the sales by the number of shelves when obtaining the sales per
- FIG. 21 is an inter-shelf distance table (FD) that stores the distance between two shelves considering an obstacle, and shows the relationship between two shelf IDs and the distance.
- FD inter-shelf distance table
- a case ID (FD1) is an ID for identifying a case.
- the date and time (FD2) is the date that is the object of simulation.
- the shelf ID 1 (FD 3) is a number for identifying the shelf 1.
- the shelf ID 2 (FD 4) is a number for identifying the shelf 2.
- shelf ID1 (FD3) and shelf ID2 (FD4) are divided into frontage (stage (horizontal) and column (vertical)), they may be stored so that they can be identified.
- the distance (FD5) is a distance between the shelf ID1 (FD3) and the shelf ID2 (FD4) considering an obstacle.
- the unit is meters. The distance can be obtained by using a general shortest path problem algorithm such as Dijkstra method, Bellman Ford method, and A * algorithm.
- FIG. 22 is a map table (GM) for storing icon information that needs to be displayed on the content (K). This is stored in the map database (G).
- GM map table
- Content (K) supports input by designating icons such as shelves and obstacles on the screen, and makes it easy for the user to understand by displaying icons.
- the map table (GM) shows a correspondence table between icons and maps in the content (K).
- the case ID (GM1) is an ID for identifying the case.
- the date and time (GM2) is the date on which the simulation is performed.
- the background map file (GM3) is a map file displayed on the background of the content (K).
- the shelf ID (GM4) is a number for identifying the shelf.
- the coordinate X (GM5) is an X coordinate value when the layout is viewed from the map reference point (origin).
- the coordinate Y (GM6) is a Y coordinate value when the layout is viewed from the map reference point (origin).
- the icon type (GM7) is the type of icon when displayed. 1 is a shelf, 2 is an obstacle, 3 is an initial stop, 4 is an entrance, 5 is an exit, and 6 is a cash register.
- a plurality of numbers may be assigned. For example, 4 and 5 for the entrance / exit.
- the region size X (GM8) indicates the size in the X-axis direction with the coordinate X (GM5) as the center when viewed from the map.
- the region size Y indicates the size in the Y-axis direction with the coordinate Y (GM6) as the center when viewed from the map.
- the takeout direction indicates the direction of the takeout port as viewed from the reference point when the shelf is installed. 1 is up, 2 is down, 3 is left, and 4 is right.
- FIG. 23 is a stop-by rate table (HT) for storing the drop-in rate for each shelf obtained by actual measurement. This is stored in the stop-by database (H).
- HT stop-by rate table
- the case ID (HT1) is an ID for identifying the case.
- the date and time (HT2) is the date on which the simulation is performed.
- the shelf ID (HT3) is a number for identifying the shelf.
- the stop-by rate (HT4) is a stop-by rate for each shelf determined by actual measurement.
- a measurement method a general measurement method such as a questionnaire, a camera, laser measurement, or a sensor can be used.
- FIG. 24 shows basic data for charging by recording the usage time and number of times of simulation, and is recorded in the charging table (IK). This is stored in the accounting database (I). Next, the contents will be described.
- a user ID (IK1) is an ID for identifying a user (US) who uses this application.
- the case ID (IK2) is an ID for identifying the case.
- the date and time (IK3) is the date on which the simulation is performed.
- the number of clicks (IK4) is the number of times a query is transmitted from the client (CL) to the application server (AS).
- the cloud usage time (IK5) stores the time taken for processing by the application server (AS). If the number of clicks (IK4) and the cloud usage time (IK5) are classified in detail, the usage status may be stored in the billing table (IK) with details for each query and page.
- the shelf coordinate information (coordinate X (GM5) and coordinate Y (GM6)) in the store, the shelf number of the shelf ( Shelf ID (GM4)) and input unit (transmission / reception unit ASS) for inputting information (map table (GM)) associated therewith, coordinate information of shelf, shelf number of shelf, and association thereof
- the first process for obtaining the stay position or stay probability at a certain time t of the customer of the store and the second process for obtaining the stay position or stay probability at the customer time (t + ⁇ t) are performed a plurality of times. By repeating, it is easy to stop by the customer for each shelf, or a simulator unit (customer simulator AP) for calculating the sales forecast for each shelf, Display unit for displaying on the expected (display CLID), and having a.
- shelf number shelf ID (FT3)
- product information product ID (FT5)
- FT5 shelf product table (FT)
- Sales information sales amount (EU5)
- product information product ID (EU1)
- information that links them sales table (EU)
- each invention according to the present embodiment is a system that can be applied to a place where people travel, and can be applied not only to stores but also to factories, construction sites, and distribution warehouses.
- AS application server ASS transceiver unit ASC control unit ASCC communication control ASCW Web server AP customer simulator APA store layout evaluation APB drop-in simulation APC store layout evaluation learning APD product shelf replacement calculation APE billing ASM storage unit D simulation database E sales database F shelf database G Map database H Stopping database I Billing database CL Client CLS Transmission / reception unit CLC Control unit CLCC Communication control CLCA Content generation CLCP Drawing setting CLCT Analysis condition CLCW Web browser CLM Storage unit CLMP Analysis condition information CLMT Drawing setting information CLI Input / output unit CLID Display CLIK Keyboard CLIM Mouse CLIU External I / O K Content NW Net Over click US user
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
「座標Pの商品Mが購入される可能性」=f(「場所の特性」、「商品の特性」、「顧客の行動特性」) …(1)
ここで、係る方法は当然、個々の担当者の技量に強く依存するものであり、複数の要因からなる複合的な系について、必ずしも妥当な配置ができるとは限らない。そこで、複合的な各要因を単純化し、商品配置の決定をより容易ならしめるための技術が望まれる。しかしながら、係る技術については、上記各特許文献はもとより、いずれの文献にも記載も示唆も無かった。
「座標Pの商品Mが購入される可能性」=「場所の特性」*g(「商品の特性」) …(2)
上記式(1)を式(2)へ変形する方法とは、顧客の移動経路情報に基づいて顧客の行動特性を定数化し、場所の特性をシミュレーションして定量的な値として算出することである。係る変形ができれば、商品配置の決定者が、場所の特性のみ、商品の特性のみをそれぞれ分けて検討することが可能となり、より商品配置の決定が容易となる。
図2A、図2B、は一つの実施形態である顧客シミュレータシステムの構成要素を示す説明図であり、図示の都合上分割して示してあるが、各々図示された各処理は相互に連携して実行される。
図2Aに示すアプリケーションサーバ(AS)は、顧客シミュレータの処理を実行する。アプリケーションサーバ(AS)において、図2Bに示すクライアント(CL)からの依頼を受けた際、設定された時刻に自動、又は、手動にて、アプリケーションが起動される。アプリケーションサーバ(AS)によって解析された結果は、図2Bに示すクライアント(CL)にネットワーク(NW)を通して送信される。
入出力部(CLI)は、ユーザとのインタフェースとなる部分である。入出力部(CLI)は、ディスプレイ(CLID)、キーボード(CLIK)およびマウス(CLIM)等を備える。必要に応じて外部入出力(CLIU)に他の入出力装置を接続することもできる。
Webブラウザ(CLCW)は、ユーザ(US)とのインタフェースであり、解析条件情報(CLMP)や描画設定情報(CLMT)の設定を行ない、また、アプリケーションサーバ(AS)での結果からコンテンツ生成(CLCA)によって出力された結果をWebブラウザ(CLCW)に表示する。
解析条件(CLCT)は、ユーザから入出力部(CLI)を介して指定される解析条件を受け取り、記憶部(CLM)の解析条件情報(CLMP)に記録する。ここでは、解析に用いるデータの案件や日時などの種類および解析のためのパラメータ等が設定される。クライアント(CL)は、これらの設定をアプリケーションサーバ(AS)に送信して解析を依頼し、それと並行して描画設定(CLCP)を実行する。
(ステップ1.)状態遷移を求める。
ここで、tr(i、j)は状態遷移確率、dd(i、j)は棚間距離テーブル(FD)、betaは移動距離(DP8)である。
(ステップ2.)正規化をする。
(ステップ1.)初期条件を決める。
ここで、pm(j、1)はホッピング別の確率テーブルである。1はホッピング回数の1番目、すなわちスタートである。なお、店舗に複数の入り口がある場合でも、この影響は正規化により吸収されるため、複数の入り口それぞれの重みを1として良い。
(ステップ2.)ホッピング数がk回のとき棚jに行く確率を決める。
ここで、tr(i、j)は状態遷移確率、pm(j、k)はホッピング別の確率テーブルである。
(ステップ3.)継続時間パラメータを決める。
(ステップ4.)2~3を総ホッピング数まで繰り返す。ホッピング別確率(APB5)の式は一例であり、他の計算式でもかまわない。
(ステップ1.)初期条件として0を代入する。
(ステップ2.)以下の式により、ホッピング数がk回までに棚jに行く確率を求める。
ここで、cc(i、k)は累積確率である。すなわち、(ホッピング数がk-1回までに棚jに立寄った確率)+(ホッピング数がk-1回までに棚jに立寄っていない確率)*(ホッピング数k回の時に棚jに立寄る確率)である。
(ステップ3.)2を総ホッピング数まで繰り返す。
(ステップ4.)ccが総ホッピング数の時の値を立寄率S(APB8)として出力する。
立寄率S(APB8)はホッピング累積確率(APB7)によって出力された結果である。詳細は図15で説明する。
商品効果(APA10)はバイアス処理(APA8)の出力結果であり、場所の効果を排除した商品群の効果を示したものである。
顧客単価=B’*a+b
a=売上モデルの傾き
b=売上モデルの切片
B’=置き換えた場所の場所バイアス
これは、一例であり、売上モデルに相応しい、他の計算式でもかまわない。
図12は、顧客シミュレータ(AP)に必要なパラメータを格納するパラメータテーブル(DP)である。
図20~21において、棚データベース(F)に格納されるテーブルを説明する。図20は、商品と棚とを関係付けることを示した、棚商品テーブル(FT)である。これを用いることによって、どの商品がどの棚に配置されているかがわかる。
距離(FD5)は、障害物を考慮した棚ID1(FD3)と棚ID2(FD4)の距離である。単位はメートルである。距離の求め方は、ダイクストラ法、ベルマンフォード法、A*アルゴリズムなど、一般的な最短経路問題のアルゴリズムを使うことができる。
ASS 送受信部
ASC 制御部
ASCC 通信制御
ASCW Webサーバ
AP 顧客シミュレータ
APA 店舗レイアウト評価
APB 立寄りシミュレーション
APC 店舗レイアウト評価学習
APD 商品棚置換計算
APE 課金
ASM 記憶部
D シミュレーションデータベース
E 売上データベース
F 棚データベース
G マップデータベース
H 立寄データベース
I 課金データベース
CL クライアント
CLS 送受信部
CLC 制御部
CLCC 通信制御
CLCA コンテンツ生成
CLCP 描画設定
CLCT 解析条件
CLCW Webブラウザ
CLM 記憶部
CLMP 解析条件情報
CLMT 描画設定情報
CLI 入出力部
CLID ディスプレイ
CLIK キーボード
CLIM マウス
CLIU 外部入出力
K コンテンツ
NW ネットワーク
US ユーザ
Claims (15)
- 顧客が店舗に入店後の時間経過に対する滞在確率に関する第1の情報、前記店舗に設けられた複数の棚のそれぞれの間の距離を示す第2の情報、および、前記店舗に顧客が滞在する滞在時間と前記顧客が前記棚間で移動を行う移動周期を示す第3の情報とが入力される入力部と、
前記第1の情報、前記第2の情報、並びに、前記第3の情報、および、
a)前記顧客は、前記店舗の入り口から移動を開始する、
b)前記顧客は、前記複数の棚のうち移動距離が遠くのより近くにある棚に移動する確率が高い、
c)前記顧客は、前記滞在時間しか前記店舗に滞在しない、
d)前記顧客は、移動先の棚をランダムに移動する、
というシミュレーション条件を記憶する記憶部と、
前記第1の情報、前記第2の情報、前記第3の情報、および、前記シミュレーション条件とを用いて、前記複数の棚のそれぞれに対して前記顧客が滞在する確率を計算するシミュレータ部と、
前記確率を前記棚に対応付けて表示する表示部と、を有することを特徴とする情報処理システム。 - 請求項1において、
前記第1の情報は、前記顧客が前記店舗に滞在する時間のオフセットと、前記顧客が前記店舗に滞在する確率が1/eになる時間の2つの値により一意に定まる関数であること特徴とする情報処理システム。 - 請求項1において、
前記シミュレータ部は、前記移動周期毎の移動のうち、t回目(tは自然数)の位置から、(t+1)回目の位置へ移動する確率を求めることを特徴とする情報処理システム。 - 請求項1において、
前記入力部に、前記店舗における売上を含む第4の情報がさらに入力され、
前記シミュレータ部は、前記第4の情報をさらに用いて、前記売上から前記確率を排除した情報である商品効果を計算し、
前記表示部は、前記確率および前記商品効果を表示することを特徴とする情報処理システム。 - 請求項4において、
前記シミュレータ部は、前記確率および前記商品効果の平均値を算出し、
前記表示部は、前記複数の棚のそれぞれを、前記確率が平均値より大きいか否か、および、前記商品効果が平均値より大きいか否か、の2つの情報を識別できる表示態様で表示することを特徴とする情報処理システム。 - 請求項4において、
前記表示部において、前記複数の棚のそれぞれは、移動可能に構成され、
前記シミュレータ部は、前記複数の棚のいずれかを移動させた際の、前記確率および前記商品効果をさらに算出することを特徴とする情報処理システム。 - 請求項1において、
前記シミュレータ部は、前記顧客の顧客単価、顧客購買点数、又は顧客購買品目数のうち少なくとも1つをさらに算出し、
前記表示部は、前記顧客の顧客単価、前記顧客購買点数、又は前記顧客購買品目数のうち少なくとも1つをさらに表示することを特徴とする情報処理システム。 - 顧客が店舗に入店後の時間経過に対する滞在確率に関する第1の情報、前記店舗に設けられた複数の棚のそれぞれの間の距離を示す第2の情報、および、前記店舗に顧客が滞在する滞在時間と前記顧客が前記棚間で移動を行う移動周期を示す第3の情報を入力として受け付ける第1の工程と、
前記第1の情報、前記第2の情報、並びに、前記第3の情報、および、
a)前記顧客は、前記店舗の入り口から移動を開始する、
b)前記顧客は、前記複数の棚のうち移動距離が遠くのより近くにある棚に移動する確率が高い、
c)前記顧客は、前記滞在時間しか前記店舗に滞在しない、
d)前記顧客は、移動先の棚をランダムに移動する、
というシミュレーション条件とを用いて、前記複数の棚のそれぞれに対して前記顧客が滞在する確率を計算する第2の工程と、
前記確率を前記棚に対応付けて表示する第3の工程と、を有することを特徴とする情報処理方法。 - 請求項8において、
前記第1の情報は、前記顧客が前記店舗に滞在する時間のオフセットと、前記顧客が前記店舗に滞在する確率が1/eになる時間の2つの値により一意に定まる関数であること特徴とする情報処理方法。 - 請求項8において、
前記第2の工程において、前記移動周期毎の移動のうち、t回目(tは自然数)の位置から、(t+1)回目の位置へ移動する確率を求めることを特徴とする情報処理方法。 - 請求項8において、
前記第1の工程において、前記店舗における売上を含む第4の情報がさらに入力され、
前記第2の工程において、前記第4の情報をさらに用いて、前記売上から前記確率を排除した情報である商品効果をさらに計算し、
前記第3の工程において、前記確率および前記商品効果を表示することを特徴とする情報処理方法。 - 請求項8において、
前記第2の工程において、前記確率および前記商品効果の平均値を算出し、
前記第3の工程において、前記複数の棚のそれぞれを、前記確率が平均値より大きいか否か、および、前記商品効果が平均値より大きいか否か、の2つの情報を識別できる表示態様で表示することを特徴とする情報処理方法。 - 請求項8において、
前記第2の工程において、前記顧客の顧客単価、顧客購買点数、又は顧客購買品目数のうち少なくとも1つをさらに算出し、
前記第3の工程において、前記顧客の顧客単価、前記顧客購買点数、又は前記顧客購買品目数のうち少なくとも1つをさらに表示することを特徴とする顧客シミュレータシステム。 - 店舗における棚の座標情報、前記棚の棚番号、および、これらを紐付けた情報を入力する入力部と、
前記棚の座標情報、前記棚の棚番号、および、これらを紐付けた情報を用いて、前記店舗の顧客のある時刻tにおける滞在位置または滞在確率を求める第1の処理と、前記顧客の時刻(t+Δt)における前記滞在位置または前記滞在確率を求める第2の処理と、を複数回繰り返すことによって、前記棚毎の前記顧客の立寄りやすさ、または、前記棚毎の売上予測を算出するシミュレータ部と、
前記立寄りやすさ、または、前記売上予測を表示する表示部と、を有することを特徴とする情報処理システム。 - 請求項14において、
前記入力部は、
前記棚番号、当該棚番号を有する棚に配置される商品情報、並びに、これらを紐付ける情報、及び、前記売上情報、前記商品情報、および、これらを紐付ける情報と、をさらなる入力とすることを特徴とする情報処理システム。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014551853A JP5905124B2 (ja) | 2013-10-25 | 2013-10-25 | 情報処理システムおよび情報処理方法 |
PCT/JP2013/078894 WO2015059807A1 (ja) | 2013-10-25 | 2013-10-25 | 情報処理システムおよび情報処理方法 |
US14/417,585 US20160005052A1 (en) | 2013-10-25 | 2013-10-25 | Information processing system and information processing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2013/078894 WO2015059807A1 (ja) | 2013-10-25 | 2013-10-25 | 情報処理システムおよび情報処理方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015059807A1 true WO2015059807A1 (ja) | 2015-04-30 |
Family
ID=52992444
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/078894 WO2015059807A1 (ja) | 2013-10-25 | 2013-10-25 | 情報処理システムおよび情報処理方法 |
Country Status (3)
Country | Link |
---|---|
US (1) | US20160005052A1 (ja) |
JP (1) | JP5905124B2 (ja) |
WO (1) | WO2015059807A1 (ja) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016199405A1 (ja) * | 2015-06-09 | 2016-12-15 | 日本電気株式会社 | 棚割支援装置、棚割支援システム、棚割支援方法、および、記録媒体 |
JP2016218821A (ja) * | 2015-05-22 | 2016-12-22 | 由紀貞 深谷 | 販売情報利用装置、販売情報利用方法、およびプログラム |
WO2018008203A1 (ja) * | 2016-07-05 | 2018-01-11 | パナソニックIpマネジメント株式会社 | シミュレーション装置、シミュレーションシステム、及びシミュレーション方法 |
CN109982251A (zh) * | 2019-04-01 | 2019-07-05 | 北京泰迪熊移动科技有限公司 | 一种驻场/离场的状态检测方法、设备以及计算机存储介质 |
JP2019113889A (ja) * | 2017-12-20 | 2019-07-11 | ヤフー株式会社 | 算出装置、算出方法、及び算出プログラム |
JP2019125312A (ja) * | 2018-01-19 | 2019-07-25 | 富士通株式会社 | シミュレーションプログラム、シミュレーション方法およびシミュレーション装置 |
JP2020095404A (ja) * | 2018-12-11 | 2020-06-18 | 東京電力ホールディングス株式会社 | 情報処理方法、プログラム、情報処理装置及び学習済みモデルの生成方法 |
JP2020205127A (ja) * | 2018-12-11 | 2020-12-24 | 東京電力ホールディングス株式会社 | 情報処理方法、プログラム及び情報処理装置 |
JP7279216B1 (ja) | 2022-01-31 | 2023-05-22 | Kddi株式会社 | 情報処理装置、情報処理方法及びプログラム |
WO2023175700A1 (ja) * | 2022-03-15 | 2023-09-21 | 日本電気株式会社 | 情報処理システム、情報処理方法、およびプログラム |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108681918A (zh) * | 2018-05-07 | 2018-10-19 | 北京学之途网络科技有限公司 | 获取顾客停留时长与商品销量关系的方法、装置和系统 |
US20200012999A1 (en) * | 2018-07-03 | 2020-01-09 | Baidu Usa Llc | Method and apparatus for information processing |
JP7127763B2 (ja) * | 2018-10-01 | 2022-08-30 | 株式会社野村総合研究所 | 解析装置 |
US20210201384A1 (en) * | 2019-12-25 | 2021-07-01 | Beijing Sensetime Technology Development Co., Ltd. | Region arrangement methods, apparatuses and storage media |
CN111125288A (zh) * | 2019-12-25 | 2020-05-08 | 北京市商汤科技开发有限公司 | 区域部署方法、装置及存储介质 |
CN114463073B (zh) * | 2022-04-08 | 2022-07-05 | 深圳市瓴风科技有限公司 | 基于用户行为的样品展示售卖方法及系统 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08166974A (ja) * | 1994-12-13 | 1996-06-25 | Kao Corp | 店舗の設計支援システム |
JP2004151955A (ja) * | 2002-10-30 | 2004-05-27 | Toppan Printing Co Ltd | 商品陳列シミュレーションサーバ装置 |
JP2013210870A (ja) * | 2012-03-30 | 2013-10-10 | Hitachi Solutions Ltd | 動線情報計測システムおよび方法、並びに情報処理装置 |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8189926B2 (en) * | 2006-12-30 | 2012-05-29 | Videomining Corporation | Method and system for automatically analyzing categories in a physical space based on the visual characterization of people |
US9245276B2 (en) * | 2012-12-12 | 2016-01-26 | Verint Systems Ltd. | Time-in-store estimation using facial recognition |
WO2015040661A1 (ja) * | 2013-09-19 | 2015-03-26 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカ | 情報端末にマーチャンダイジング情報を表示させる制御方法 |
-
2013
- 2013-10-25 JP JP2014551853A patent/JP5905124B2/ja active Active
- 2013-10-25 WO PCT/JP2013/078894 patent/WO2015059807A1/ja active Application Filing
- 2013-10-25 US US14/417,585 patent/US20160005052A1/en not_active Abandoned
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08166974A (ja) * | 1994-12-13 | 1996-06-25 | Kao Corp | 店舗の設計支援システム |
JP2004151955A (ja) * | 2002-10-30 | 2004-05-27 | Toppan Printing Co Ltd | 商品陳列シミュレーションサーバ装置 |
JP2013210870A (ja) * | 2012-03-30 | 2013-10-10 | Hitachi Solutions Ltd | 動線情報計測システムおよび方法、並びに情報処理装置 |
Non-Patent Citations (1)
Title |
---|
TAKEHIKO ABE: "Layout Support System for Retail Store Using Multi-agent", PROCEEDINGS OF THE 2006 IEICE GENERAL CONFERENCE JOHO SYSTEM 1, 8 March 2006 (2006-03-08), pages 89 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2016218821A (ja) * | 2015-05-22 | 2016-12-22 | 由紀貞 深谷 | 販売情報利用装置、販売情報利用方法、およびプログラム |
JP2021108209A (ja) * | 2015-06-09 | 2021-07-29 | 日本電気株式会社 | 棚割支援装置、棚割支援システム、棚割支援方法、および、記録媒体 |
JP7501605B2 (ja) | 2015-06-09 | 2024-06-18 | 日本電気株式会社 | 棚割支援装置、棚割支援方法、および、プログラム |
JPWO2016199405A1 (ja) * | 2015-06-09 | 2018-04-12 | 日本電気株式会社 | 棚割支援装置、棚割支援システム、棚割支援方法、および、プログラム |
WO2016199405A1 (ja) * | 2015-06-09 | 2016-12-15 | 日本電気株式会社 | 棚割支援装置、棚割支援システム、棚割支援方法、および、記録媒体 |
JP7205565B2 (ja) | 2015-06-09 | 2023-01-17 | 日本電気株式会社 | 棚割支援装置、棚割支援システム、棚割支援方法、および、記録媒体 |
JPWO2018008203A1 (ja) * | 2016-07-05 | 2018-09-27 | パナソニックIpマネジメント株式会社 | シミュレーション装置、シミュレーションシステム、及びシミュレーション方法 |
WO2018008203A1 (ja) * | 2016-07-05 | 2018-01-11 | パナソニックIpマネジメント株式会社 | シミュレーション装置、シミュレーションシステム、及びシミュレーション方法 |
JP2019113889A (ja) * | 2017-12-20 | 2019-07-11 | ヤフー株式会社 | 算出装置、算出方法、及び算出プログラム |
JP2019125312A (ja) * | 2018-01-19 | 2019-07-25 | 富士通株式会社 | シミュレーションプログラム、シミュレーション方法およびシミュレーション装置 |
US11120386B2 (en) | 2018-01-19 | 2021-09-14 | Fujitsu Limited | Computer-readable recording medium, simulation method, and simulation apparatus |
JP2020095404A (ja) * | 2018-12-11 | 2020-06-18 | 東京電力ホールディングス株式会社 | 情報処理方法、プログラム、情報処理装置及び学習済みモデルの生成方法 |
JP2020205127A (ja) * | 2018-12-11 | 2020-12-24 | 東京電力ホールディングス株式会社 | 情報処理方法、プログラム及び情報処理装置 |
CN109982251A (zh) * | 2019-04-01 | 2019-07-05 | 北京泰迪熊移动科技有限公司 | 一种驻场/离场的状态检测方法、设备以及计算机存储介质 |
JP7279216B1 (ja) | 2022-01-31 | 2023-05-22 | Kddi株式会社 | 情報処理装置、情報処理方法及びプログラム |
JP2023111330A (ja) * | 2022-01-31 | 2023-08-10 | Kddi株式会社 | 情報処理装置、情報処理方法及びプログラム |
WO2023175700A1 (ja) * | 2022-03-15 | 2023-09-21 | 日本電気株式会社 | 情報処理システム、情報処理方法、およびプログラム |
Also Published As
Publication number | Publication date |
---|---|
US20160005052A1 (en) | 2016-01-07 |
JPWO2015059807A1 (ja) | 2017-03-09 |
JP5905124B2 (ja) | 2016-04-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5905124B2 (ja) | 情報処理システムおよび情報処理方法 | |
US11966945B2 (en) | Conducting multivariate experiments | |
Sarkar | An inventory model with reliability in an imperfect production process | |
US11856258B2 (en) | Systems and methods for spatial and temporal experimentation on content effectiveness | |
US8600924B2 (en) | Method and system to manage multiple party rewards using a single account and artificial intelligence | |
Ghasemkhani et al. | Integrated production-inventory-routing problem for multi-perishable products under uncertainty by meta-heuristic algorithms | |
CN102436625A (zh) | 在地图上显示产品推荐 | |
TW201303784A (zh) | 準則式食物購買管理 | |
US20130024270A1 (en) | Systems and methods for transactions-based content management on a digital signage network | |
US20190122233A1 (en) | Simulation device, simulation system, and simulation method | |
JP6906430B2 (ja) | 情報処理装置、情報処理方法および情報処理プログラム | |
US20190213616A1 (en) | Shopper Traffic Flow Spatial Analytics Based on Indoor Positioning Data | |
Goncalves et al. | Crowdsourcing queue estimations in situ | |
Suher et al. | The power of atlas: Why in-store shopping behavior matters | |
Torrens | Agent models of customer journeys on retail high streets | |
JP7037747B2 (ja) | 施策策定装置、施策策定方法および施策策定プログラム | |
US20140052502A1 (en) | Balanced web analytics scorecard | |
JP7066958B2 (ja) | 訪問先推薦装置、訪問先推薦方法および訪問先推薦プログラム | |
JP5900154B2 (ja) | 商品推薦方法及びサーバ装置 | |
King et al. | Simulating the effect of measurement errors on pedestrian destination choice model calibration | |
JP2003223548A (ja) | 顧客行動解析方法および解析プログラム | |
US20210326910A1 (en) | System and method for optimizing an observation campaign in response to observed real-world data | |
JP2011081451A (ja) | 表示データ生成装置及びプログラム | |
JP2011090385A (ja) | 店舗内案内装置および店舗内案内方法 | |
Cain | Brand management and the marketing mix model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2014551853 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14417585 Country of ref document: US |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13896076 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13896076 Country of ref document: EP Kind code of ref document: A1 |