JP2010079808A - Circulation point sorting system and circulation point sorting method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To sort stores included in each delivery course in order to create a plurality of delivery courses for a plurality of circulation points so that objects can efficiently be delivered even when the number of delivery days a week and deliverable days of the week are variously different. <P>SOLUTION: A circulation point sorting system (store sorting system 100) sorts stores to be included in each delivery course. The store sorting system 100 includes a circulation point information storing part (store information storing part T3) for storing the number of delivery days to each store a week, deliverable days of the week and position information of the store by store, a day of the week allocation processing part 106 for allocating delivery days for delivering objects to the store by store, and a store sorting part 122 for sorting stores to be included in each delivery course in order to create delivery courses that respectively include stores to which each day of the week is allocated as a delivery day of the week by the day of the week allocation processing part 106 by day of the week. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a traveling point classification system and a traveling point classification method.

Patent Document 1 (Japanese Patent Application Laid-Open No. 2004-127038) describes a system that divides an area where a plurality of vending machines are installed into areas in charge of workers who perform patrol work of vending machines. The system includes an initial setting unit for setting a plurality of areas including at least one vending machine in accordance with the number of persons in charge in a region where a plurality of vending machines are installed, and vending machines included in each area. A calculation unit that calculates a statistic determined by the characteristics of the image, and after selecting a specific area in consideration of the statistic, a process of adding a vending machine to the specific area and expanding the area is repeatedly executed. A processing unit that terminates the process when the condition is satisfied, and determines the obtained area as a responsible area of the plurality of responsible persons. Thereby, the area in which a plurality of vending machines are installed can be divided into areas in charge so that each worker can patrol efficiently.
JP 2004-127038 A

  In the technique described in Patent Literature 1, a course that can be visited efficiently is calculated by calculating a patrol order in which a store included in the course can be visited on the shortest route using an approximation algorithm. However, the traveling points of a plurality of stores and the like include those having different delivery days per week and deliverable days of the week. Conventionally, in such delivery to various traveling points, there has been a problem in that a delivery day for each traveling point is appropriately allocated and a delivery course for each day is efficiently created. For example, when the number of delivery days per week is 1 day and there are 5 delivery days of the week, the delivery day of the week to the circulation point is 5 days of delivery day according to the delivery status to the other circulation points. It can be set freely from the inside. Conventionally, a plurality of delivery courses for each day of the week that can be efficiently circulated while appropriately assigning the delivery days to the tour points where delivery days can be set freely and maintaining the number of delivery courses for each day of the week appropriately. There was a problem in classifying the stores to be included in each delivery course for creation.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to efficiently deliver articles to a plurality of circulation points even when the number of delivery days per week and the day of delivery are different. An object of the present invention is to provide a technique for classifying stores to be included in each delivery course in order to create a plurality of delivery courses.

According to the present invention,
A system for classifying the circulation points to be included in each delivery course in order to create a plurality of delivery courses for delivering articles to a plurality of circulation points included in a predetermined area,
A traveling point information storage unit for storing the number of days of delivery per week and the day of delivery available to the traveling point, and the position information of the traveling point, for each traveling point;
A day-of-week assignment processing unit for assigning a delivery day of the week for delivering the article to the circulation point for each of the circulation points;
For each day of the week, a traveling point classifying unit that classifies the traveling points to be included in each delivery course in order to create a delivery course that includes each of the traveling points assigned by the day of the week allocation processing unit as the delivery day of the week;
Including
The day of week allocation processing unit
A day of week allocation unit for fixing a delivery day of the week to the patrol point where the delivery day of the week can be fixed based on the number of delivery days and the day of delivery;
A key tour point selection unit that selects a plurality of tour points as key tour points so as to be distributed within the area based on position information of each tour point from among the tour points whose delivery day has already been fixed. When,
A plurality of clusters including each of the key circulation points, wherein the delivery day of the week fixed to the key circulation point is fixed as a delivery day of the week, and the circulation points that can be delivered to each of the delivery days are sequentially assigned to each of the clusters. A cluster creation unit that performs processing to add and classifies the traveling points included in the area into any of the clusters;
In each of the plurality of clusters, when the delivery days fixed to the cluster include a plurality of days, a group for each day is created, and the delivery day is already fixed among the tour points included in the cluster. The distribution day of the week is not yet fixed so that the difference in the statistics calculated based on the tour points included in the group of the day of week is reduced. A group creation unit that sequentially adds the traveling points included in the cluster to any day group,
Including
The day of week allocation unit is configured to fix the tour day in which the delivery day of the week can be fixed based on the delivery day of the week that is created by the cluster creation unit and the delivery days of each tour point and the day of delivery that can be delivered. There is provided a traveling point classification system for fixing a day of a delivery to a point and fixing a day of the group as a day of the delivery to the traveling point included in the group of each day of the week created by the group creation unit.

According to the present invention,
A traveling point classification method for classifying traveling points to be included in each delivery course in order to create a plurality of delivery courses for delivering articles to a plurality of traveling points included in a predetermined area,
For each traveling point, reading data from a traveling point information storage unit that stores the number of delivery days per week to the traveling point and the day that can be delivered, and location information of the traveling point;
Assigning a delivery day for delivering the article to the circulation point for each of the circulation points;
For each day of the week, in the step of assigning the delivery day of the week, the step of categorizing the circulation points to be included in each delivery course in order to create a delivery course that includes each of the tour points assigned as the delivery day of the week;
Including
The step of assigning the delivery day includes
Fixing the delivery day of the week to the patrol point where the delivery day can be fixed based on the delivery days and the deliverable day of the week for each round point;
Selecting a plurality of the tour points as key tour points so as to be distributed in the area based on the location information of the visit points from the tour points where the delivery day is already fixed;
A plurality of clusters including each of the key circulation points, wherein the delivery day of the week fixed to the key circulation point is fixed as a delivery day of the week, and the circulation points that can be delivered to each of the delivery days are sequentially assigned to each of the clusters. Performing a process of adding, and classifying the cyclic points included in the area into any of the clusters;
In each of the plurality of clusters, when the delivery days fixed to the cluster include a plurality of days, a group for each day is created, and the delivery day is already fixed among the tour points included in the cluster. The distribution day of the week is not yet fixed so that the difference in the statistics calculated based on the tour points included in the group of the day of week is reduced. Sequentially classifying the tour points included in the cluster into any of the day groups,
Including
The step of fixing a delivery day of the week to the round point includes the day of delivery fixed to each of the plurality of clusters created in the step of classifying into the cluster, the number of days of delivery of the round point, and the day of delivery available. Based on the day of the group, the day of delivery of the group is fixed as the day of delivery to the rounding point included in the group of the day of week created in the step of classifying the group into the group. A cyclic point classification method for fixing the

  With such a configuration, in order to create a plurality of delivery courses that can efficiently deliver articles to a plurality of circulation points even when the number of delivery days per week and the deliverable days of the week are different. The traveling points included in each delivery course can be classified. Here, the traveling point can be, for example, a store such as a convenience store or a supermarket, or a vending machine.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

  According to the present invention, even when the delivery days per week and the deliverable days of the week are different, each delivery course is used to create a plurality of delivery courses capable of efficiently delivering articles to a plurality of circulation points. Stores to be included.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same reference numerals are given to the same components, and the description will be omitted as appropriate.

  Hereinafter, an embodiment of the present invention will be described by taking as an example a case where the traveling point is a store. The store can be, for example, a convenience store or a supermarket. In the present embodiment, the traveling point classification system may be a store classification system. In the present embodiment, the traveling point classification system classifies stores to be included in each delivery course in order to create a delivery course that delivers articles from a base to a plurality of stores (circulating points) and returns to the base. These stores include stores having different delivery days per week and deliverable days of the week.

FIG. 1 is a flowchart showing an example of a processing procedure of a store (traveling point) classification method in the present embodiment.
First, the number of delivery courses for each day of the week required in this area is calculated (step S10). Subsequently, the delivery day of the week to each store is determined (step S20). Next, for each day of the week, the stores to be included in each delivery course are classified in order to create a delivery course that includes stores for which the day of the week is determined as the delivery day in step S20 (step S30).

  FIG. 2 is a block diagram showing a configuration of the store classification system 100 in the present embodiment.

The store classification system 100 includes a display processing unit 102, a condition setting receiving unit 104, a day of the week allocation processing unit 106, a calculation unit 120, a store classification unit (traveling point classification unit) 122, a required number calculation processing unit 124, and a map information storage unit T1. , A base information storage unit T2, a store information storage unit T3, a condition setting storage unit T4, a course information storage unit T5, a logic storage unit T6, and a temporary storage unit T7. The day-of-week assignment processing unit 106 includes a pattern assignment unit 108, a key store selection unit (key circulation point selection unit) 110, a cluster creation unit 112, a day-of-week assignment unit 114, a group creation unit 116, and an adjustment processing unit 118.
The processing of the necessary number calculation processing unit 124 will be described later.

  The day-of-week assignment processing unit 106 performs a process of determining a delivery day of the week to each store in step S20 of FIG. FIG. 3 is a flowchart showing in detail the processing procedure of the day-of-week assignment processing unit 106 corresponding to step S20 of FIG.

  First, the day-of-week assignment unit 114 fixes the delivery day of the store that can fix the delivery day of the week based on the number of delivery days to each store and the deliverable day of the week (step S100). Here, being able to fix the delivery day means that the delivery day is uniquely determined. Next, the pattern assigning unit 108 assigns a basic day pattern for stores to which a predetermined basic day pattern can be assigned among stores whose delivery days are not fixed yet (step S102). Here, the day-of-week assignment unit 114 fixes the delivery day of the store that can fix the delivery day of the week based on the basic day-of-week pattern assigned to each store, the number of delivery days to each store, and the deliverable day of the week (step S104).

  After that, the key store selection unit 110 selects a plurality of stores having a fixed delivery day as a key store from a plurality of stores in order to fix the delivery day of the store whose delivery day is not yet fixed ( Step S106). Here, the key store selection unit 110 selects a plurality of stores as key stores so as to be distributed in the area based on the position information of each store.

  Subsequently, the cluster creating unit 112 creates a plurality of clusters including each key store, the delivery day fixed to the key store being fixed as the delivery day, and stores that can be delivered to each cluster on each delivery day. Processing for sequentially adding is performed, and stores included in the area are classified into any cluster (step S108). Thereafter, based on the delivery day assigned to the cluster including each store, the day of week assignment unit 114 fixes the delivery day of the store where the delivery day can be fixed among the stores whose delivery day is not yet fixed (step S110). ).

  The group creation unit 116 creates a group for each of a plurality of days of the week when a delivery day fixed to the cluster includes a plurality of days among the plurality of clusters, and the difference in the statistics of the stores included in each group Stores are sequentially added so as to become smaller (step S112). The day allocation unit 114 fixes the day of the group as the delivery day of the store included in each of the plurality of groups created by the group creation unit 116 (step S114). As described above, the delivery day of the week to each store in the area is fixed.

  Returning to FIG. 2, the calculation unit 120 delivers goods to the stores included in each delivery course at the time of travel classification between the stores, the travel time between the stores, the average value of these, and the store classification. Necessary course time, total amount when goods are delivered to stores included in each delivery course at the time of store classification, etc. are calculated. Note that the calculation unit 120 calculates the time so as to be the optimum road travel time as the travel time between the base and the store and between the stores.

  Further, the calculation unit 120 can calculate a tour order in which these stores can be visited on the shortest route based on the location information of stores included in the same course. In an area including a plurality of stores, the shortest route for visiting these stores can be obtained by a so-called traveling salesman problem method. The traveling salesman problem is a problem of searching for the shortest route when a salesman travels through n cities once. If the number of cities is n, (n-1)! When there are combinations of / 2, and the number of n increases, the number of all combinations increases exponentially, so it is difficult to obtain an exact solution. Therefore, various approximation algorithms are used to optimize the combination. Examples of such approximate algorithms include random search, hill-climbing method, simulated annealing (SA), genetic algorithm (GA), and the like. In the present embodiment, the calculation unit 120 can calculate a tour order in which the stores included in the course can be visited on the shortest route using such an approximation algorithm as appropriate.

  Store classification part 122 performs processing which classifies a store included in each delivery course in order to create a plurality of delivery courses for every day of the week of Step S30 of FIG.

  The display processing unit 102 performs processing for displaying various data to be processed by the store classification system 100 on a screen (not shown) such as a display.

  The condition setting accepting unit 104 accepts setting of various conditions for store classification from the user. For example, the condition setting accepting unit 104 may include, for example, a daily standard work time, a lower limit work time, a work start time, an article loading time, a break time, a work time after returning to the base, The standard work time, the standard movement speed between stores, the standard movement speed between the base and the store, the maximum and minimum loading capacity of the truck, and the setting of the allowable loading ratio of the loading capacity are accepted. In addition, the condition setting receiving unit 104 can also receive setting of the upper limit and the lower limit of the course time. The condition setting storage unit T4 stores the set conditions.

  The map information storage unit T1 stores map information data of a region that is a target of store classification. The base information storage unit T2 stores base information data. The base information data includes a base identification code, a base name, a base position (latitude and longitude, or X coordinate and Y coordinate), and the like. Here, the base is a delivery base such as a distribution center in the area. In the present embodiment, a store included in a predetermined area within a predetermined range from the base is set to belong to the base. The logic storage unit T6 stores logic relating to various processing procedures. The course information storage unit T5 stores information on the course created by the store classification unit 122.

  The store information storage unit T3 stores store information data. The store information data includes a base identification code, a store identification code, a store name, a store location (latitude, longitude and height, or X coordinate, Y coordinate and Z coordinate), delivery amount per time, and one week. Delivery date and time conditions such as the number of delivery days per day and deliverable days of the week, work time per minute (minutes), travel speed from the base (km / hour), travel speed between stores (km / hour), monthly sales (yen) , Including information on stores such as monthly sales volume (individual) for each product.

  FIG. 4 is a block diagram showing an example of the configuration of the store information storage unit T3 in the present embodiment. The store information storage unit T3 includes a delivery information storage unit 126, a day of week allocation information storage unit 128, and an inter-store travel time storage unit 328. FIG. 5 is a diagram showing a part of the data structure of the delivery information storage unit 126 in the present embodiment. Here, the delivery information storage unit 126 includes a base code field, a store code field, a position field, a work time field, a delivery amount field, a delivery day (week) field, and a round-trip time (minutes) field between the base and the store. , And a day of the week that can be delivered.

  In the present embodiment, delivery is not performed on Sundays in one week, but is performed in any of Monday to Saturday. That is, the maximum number of delivery days can be 6 days. The number of delivery days “n” means that delivery is performed n days a week. For example, when the number of delivery days is “6”, it indicates that 6-day delivery is required per week. Therefore, it is necessary for a store with a delivery date of 6 days to deliver daily from Monday to Saturday. When the number of delivery days is “0.5”, it indicates that one-day delivery is required in two weeks. The day of the week that can be delivered to each store is stored in the deliverable day column. Further, here, a configuration is shown in which a deliverable day of the week column is provided, but when a delivery day of the week is designated, a configuration of a delivery designated day of the week column for storing the day of the week may be provided. Moreover, it can also be set as the structure which provided the non-delivery day column which stored the day of the week which cannot deliver to each store instead of the deliverable day column.

In the example shown in FIG. 5, for example, the store with the store code “a” belongs to the base with the base code “1000”, the position is “(X ₁ , Y ₁ , Z ₁ )”, and the working time at this store is “5 minutes”, delivery amount per delivery is “0.001t”, delivery days is “6”, round-trip travel time between bases and stores is “40 minutes”, and delivery days are “Monday Tuesday Wednesday Thursday "Sat".

  Further, for example, the store with the store code “b” has a delivery time of “5”, a round-trip travel time between the base and the store of “60 minutes”, and a deliverable day of the week is “Monday, Tuesday, Wednesday, Friday and Friday”. Similarly, the store with the store code “c” has a delivery date of “5”, a round-trip travel time between the base and the store of “40 minutes”, and a deliverable day of the week is “Monday, Wednesday, Friday and Saturday”. The store with the store code “d” has a delivery date of “4”, a round-trip travel time between the base and the store of “60 minutes”, and a deliverable day of the week is “Monday, Tuesday, Wednesday, Friday and Friday”. The store with the store code “e” has a delivery date of “4”, a round-trip travel time between the base and the store of “40 minutes”, and a deliverable day of the week is “Monday, Wednesday, Friday and Saturday”.

  FIG. 6 is a block diagram showing an example of the configuration of the condition setting storage unit T4 in the present embodiment. The condition setting storage unit T4 includes a pattern storage unit 130 and a day of the week fixed rule storage unit 132.

FIG. 7 is a diagram illustrating a part of the configuration data structure of the pattern storage unit 130.
In the present embodiment, the pattern storage unit 130 stores a basic day of the week pattern in which a delivery day is specified for each delivery day per week. Here, the basic day of the week pattern can be set so that the delivery days are dispersed as much as possible. For example, when the number of delivery days is “3”, two basic day-of-week patterns can be set: a basic day-of-week pattern 3a delivered to “Monday Wednesday” and a basic day-of-week pattern 3b delivered to “Tuesday”. . In this way, the delivery day of the week can be made even within each basic day pattern, and the delivery day of the week included in each of the two basic day patterns does not overlap each other. The delivery day of the week can be even.

  Further, when the number of delivery days is “5”, the basic day-of-week pattern is to deliver to “Monday Tuesday Wednesday Thursday”. When the number of delivery days is “4”, the basic day-of-week pattern is to deliver to “Monday, Tuesday, Thursday”. When the number of delivery days is “2”, the three basic days of the week are the basic day pattern 2a delivered to “Monday”, the basic day pattern 2b delivered to “Tuesday”, and the basic day pattern 2c delivered to “Wednesday”. A pattern is set. In this way, the delivery day of the week can be made uniform within each basic day pattern, and the delivery day of the week included in each of the three basic day patterns does not overlap with each other. The delivery day of the week can be even. When the number of delivery days is “1” or “0.5”, the basic day pattern is not set.

FIG. 8 is a diagram showing a part of the data structure of the day fixed rule storage unit 132.
In the present embodiment, the day-of-week fixing rule storage unit 132 stores a rule for the day-of-week assignment unit 114 to fix the delivery day of the week in steps S102 and S104 of FIG. 3 as an initial setting.

  Here, first, as the first condition, when “delivery days per week = number of deliverable days”, “the day of the week” is set. That is, for a store whose delivery days per week are the same as the number of deliverable days, the deliverable day is fixed as the delivery day. For example, in the case of a store having a delivery date of “2” and a deliverable day of the week being “Monday and Saturday”, “Monday and Saturday” is fixed as the delivery day of the store regardless of the basic day of the week pattern described with reference to FIG. Is done.

  Whether to fix the day of the week specified in the basic day of the week pattern as the delivery day of the store when delivery is possible on the day of the week specified in the basic day of the week pattern for each delivery day per week, When delivery is not possible on the day of the week specified by the basic day of the week pattern, whether or not to fix the delivery day of the store is set. For example, when the number of delivery days is “5”, the setting is “basic day of the week pattern if basic day pattern 5 is possible, and possible day of week when impossible”. For example, when the number of delivery days is “4”, the setting is “basic day of the week pattern if basic day pattern 4 is possible; For example, when the number of delivery days is “2”, “if only one of the basic day patterns 2a to 2c is possible, the basic day pattern, and the other day of the week is not fixed” is set. When the delivery date is “3”, “not fixed” is set.

  The pattern assignment unit 108 assigns a basic day of the week pattern to each store with reference to the delivery information storage unit 126 and the pattern storage unit 130. The pattern assignment unit 108 stores the assigned basic day of week pattern in the day of the week assignment information storage unit 128 for each store. Further, the day-of-week assignment unit 114 refers to the basic day-of-week pattern of each store assigned by the pattern assignment unit 108 and the day-of-week fixing rule storage unit 132, and fixes the delivery day of the store that can fix the delivery day of the week. When the delivery day of the week to each store is fixed, the day-of-week assignment unit 114 stores that fact in the day-of-week assignment information storage unit 128.

FIG. 9 is a diagram illustrating a part of the data structure of the day-of-week assignment information storage unit 128 included in the store information storage unit T3.
First, according to the procedure shown in FIG. 8, the day allocation unit 114 fixes the delivery day of the store where the number of delivery days = the number of deliverable days. Stores with store codes “a”, “c”, “h”, and “k” each have a delivery day = the number of deliverable days, so that day of the week is fixed as the delivery day of the week. The delivery day of the store with the store code “a” is “6 (Monday, Wednesday, Wednesday, Thursday, and Friday)”, and since the delivery day is also fixed, a flag indicating that the day of the week has been fixed is attached to the day fixed flag column. . The delivery day of the store with the store code “c” is “5 (Monday, Wednesday, Thursday, and Friday)”, and a flag indicating that the day of the week has been fixed is added to the day of the week fixed flag column. The store with the store code “h” has a delivery date of “2”, but only the basic day pattern 2a shown in FIG. 7 can be assigned. Therefore, the store with the store code “h” is assigned the basic day of the week pattern 2a, and a flag indicating that the day of the week is already fixed is added to the day of the week fixed flag column. Similarly, since the delivery day of the store with the store code “k” is “1 (month)” and the delivery day is also fixed, a flag indicating that the day of the week has been fixed is attached to the day fixed flag column.

  Next, the pattern assigning unit 108 assigns a basic day of the week pattern to stores that can be assigned a pattern. The store with the store code “b” has a delivery date of “5” and can be assigned the basic day pattern 5 shown in FIG. Therefore, the basic day pattern 5 is assigned to the store with the store code “b”. Further, the store with the store code “d” has a delivery date of “4”, and the basic day-of-week pattern 4 shown in FIG. 7 can be assigned. Therefore, the basic day pattern 4 is assigned to the store with the store code “d”.

  The store with the store code “f” has a delivery date of “3” and can be assigned the basic day-of-week pattern 3a shown in FIG. Therefore, the basic day pattern 3a is assigned to the store with the store code “f”.

  The store with the store code “g” has a delivery date of “2” and can be assigned the basic day pattern 2b or 2c shown in FIG. Therefore, the basic day pattern 2b or 2c is assigned to the store with the store code “g”.

  Further, the day allocation unit 114 refers to the day fixing rule storage unit 132 in FIG. 8 and fixes the delivery day of the store in which the delivery day can be fixed among the stores to which the basic day pattern is allocated. Here, a flag indicating that the day of the week has already been assigned is assigned to the store codes “b” and “d”.

  The store with the store code “c” has a delivery date of “5”, but the basic day pattern 5 shown in FIG. 7 cannot be assigned. On the other hand, according to the day fixing rule shown in FIG. 8, the possible day of the week is fixed as the delivery day of the week. (Monday, Wednesday, Friday and Saturday) ”is assigned. In addition, a flag indicating that the day of the week has been fixed is attached to the day of the week fixed flag column.

  On the other hand, basic day patterns cannot be assigned to stores such as store codes “e” and “j”. Therefore, no delivery day is assigned to these stores, and no fixed day flag is attached.

  This completes the initial setting in which the day of the week assigning unit 114 fixes the delivery day of the week. Next, a process for fixing the delivery day of the store where the delivery day could not be fixed in the initial setting will be described.

FIG. 10 and FIG. 11 are flowcharts showing a procedure for the key store selection unit 110 to select a key store. FIG. 10 shows the procedure of step S106 of FIG.
First, the key store selection unit 110 sets all stores whose delivery days are already fixed by the day allocation unit 114 as key store candidates (step S200). Next, based on the location information of each store, a key store candidate is added to a location where the key store candidates are insufficient (step S210). FIG. 11 is a flowchart showing this procedure.

  First, MST (Minimum Spanning Tree) is created at all stores in the area (step S300). Next, the MST is divided at an edge exceeding the maximum movement time and divided into a plurality of spare clusters (step S310). From the divided spare clusters, a spare cluster including only stores less than the predetermined minimum number of stores is deleted (step S320). In the remaining spare cluster, if a store whose delivery day is already fixed is included in the spare cluster, the spare cluster is deleted (step S330). Next, for each remaining spare cluster, the day of the week of the spare cluster is determined (step S340).

The day of the week for the spare cluster can be determined by the following procedure.
The number of weighted stores on the deliverable days of each store included in the spare cluster is totaled. Here, for stores with a delivery date of 1 or more, 1 point is added to each day of the week that can be delivered to the store. In addition, for stores with delivery days of less than 1, such as 0.5, 0.5 points are added to each day that can be delivered to the store. For example, if there are stores whose delivery days are four and the day of delivery is Monday, Thursday, Friday, Saturday, 1 point is added to the Monday, Thursday, Friday, Saturday, and Saturday respectively. In this way, the day of the week with the largest number of weighted stores in each spare cluster can be set as the day of the spare cluster. Here, when there are multiple days of the week with the same number of weighted stores, for example, the days of the week can be determined in the order of Monday, Tuesday, Wednesday, Thursday and Saturday.

  Thereafter, a store having the longest travel time between the base and the store among the stores that can be delivered on the delivery day set for each spare cluster is selected as a key store candidate of the spare cluster (step S350).

This process will be described with reference to FIGS.
FIG. 12 shows a base 200 and a plurality of stores 202 in the area.
First, the MST 204 is created at all stores in the area (FIG. 12 (a)). Next, a plurality of spare clusters 206, 208, 210, and 212 are created by dividing the MST 204 at an edge that exceeds the maximum travel time that is the maximum delivery time of one delivery course (FIG. 12B). From the divided spare clusters, spare clusters that include less than a predetermined minimum number of stores are deleted. The setting such as the minimum number of stores can be stored in the condition setting storage unit T4.

  Here, for example, when the minimum number of stores is set to “7”, the spare cluster 208 having six stores is deleted (FIG. 12C). Further, if the selected spare cluster includes a store in which the delivery day is already fixed in the spare cluster, the spare cluster is also deleted. For example, if the spare cluster 210 includes a store with the store code “a”, the delivery cluster day is already fixed for the store with the store code “a”, so the spare cluster 210 is deleted (FIG. 13A).

  Next, as described in step S340 in FIG. 11, the day of the week for each spare cluster is determined. Here, for example, it is assumed that the delivery day of the backup cluster 206 is determined as “Tuesday” and the delivery day of the backup cluster 212 is determined as “Thursday”. In that case, among the stores that can be delivered on the day of the week set in each spare cluster, the store having the longest travel time between the base 200 and the store is determined as the key store candidate of the spare cluster (FIG. 13 ( b)).

  In the above processing, if the stores included in the area are classified into a plurality of spare clusters based on the location information of each store, and the stores whose delivery day is already fixed are not included in the spare clusters, To select key store candidates. That is, a key store candidate can be set in a location where no store that is a key store candidate exists in the area. Thereby, key store candidates can be distributed and selected in the area.

  Returning to FIG. 10, the key store selection unit 110 gives priority to a store having a large number of delivery days from the key store candidates selected by the above processing, and when the delivery days are the same, A predetermined number of key stores are selected in descending order of the travel time between them (step S220). Here, the predetermined number can be a necessary number of delivery courses calculated by a necessary number calculation processing unit 124 described later.

Returning to FIG. 2, the cluster creation unit 112 creates a plurality of clusters including each key store, the delivery day fixed to the key store being fixed as the delivery day, and can be delivered to each cluster on each delivery day. Process to add stores sequentially.
Here, the cluster creation unit 112 captures stores that satisfy all of the following conditions.
(A) The delivery day of the store is already fixed,
(B) The travel time from the key store to the store is within a predetermined time,
(C) The delivery day of the week included in the delivery day fixed to the key store.

For example, the cluster creation unit 112 can acquire stores that satisfy all of the above conditions in order from the key store of each cluster. The cluster creation unit 112 can calculate the limited number of stores to be imported into each cluster, and can import the stores into each cluster until the limited number of stores is reached. The number of restricted stores can be calculated based on the following formula, for example.
Number of restricted stores = number of deliverable stores × delivery days fixed in key stores = (delivery time—average round-trip travel time between base stores + average store travel time) / (average store travel time + average store work time) × Delivery days fixed in key store

  When the number of stores included in each cluster exceeds the limit number of stores, the cluster creation unit 112 removes one store acquired last, and finishes importing stores into the cluster. Moreover, even if all the stores which can be taken in are taken in, the cluster preparation part 112 will complete | finish the taking in of the store to the cluster, when the number of stores contained in the cluster is below a limit store number. Thereafter, the cluster creation unit 112 calculates the center of the cluster based on the location information of the stores included in the cluster for each cluster, and sets the store closest to the center as the final key store of the cluster. The center of the cluster can be obtained by calculating the barycentric coordinate based on the coordinates (latitude / longitude) of the stores included in each cluster. In addition, the center of this cluster is displayed on the map coordinates (planes with x-axis and y-axis) together with the stores included in each cluster, and the store closest to the center of the cluster is set as the final key store of the cluster. be able to.

  Subsequently, the cluster creation unit 112 imports stores whose delivery days are not fixed yet into the cluster. Here, the cluster creation unit 112 has a cluster in which the number of stores included in each cluster is more than the limited number of stores, a cluster with a large number of delivery days, and a large average travel time between each store and base included in the cluster. A cluster can be preferentially selected and a store can be taken into the selected cluster. In addition, the cluster creation unit 112 preferentially captures stores with many restrictions for capturing, such as stores with many delivery days and stores with few days of delivery.

Here, the cluster creation unit 112 captures stores that satisfy all of the following conditions.
(A) The travel time from the key store to the store is within a predetermined time,
(B) The delivery day of the week included in the delivery day fixed to the key store.

  Further, in the case where only one of the stores having a plurality of delivery days is included in the delivery day fixed to the key store, the cluster creation unit 112 can capture only that day. When there are a plurality of fixed delivery days in each cluster, the cluster creation unit 112 manages the number of stores so as not to exceed the limit number of stores for each day of the week, and finishes capturing the days of the week when the limit number of stores is reached. The cluster creation unit 112 calculates the number of weighted stores in the same procedure as the process for determining the day of the week for the above-described spare cluster.

  For example, in the present embodiment, a store with a delivery date of “3” is not fixed on the day of the week specified by the basic day of the week pattern. For example, the store with the store code “s” has “Monday Wednesday” on the day of delivery. If the delivery day fixed in the key store includes Monday but does not include Wednesday and Friday, the Monday of the store with the store code “s” is taken into the cluster and the Monday is fixed. Similarly, for Wednesday and Friday, if there is a similar cluster, it is taken into that cluster. Further, for example, when there is no cluster that can be captured for Friday, for the stores that are not yet captured in any cluster, addition of the key store candidates in step S210 of FIG. 10 described above and selection of the key store in step S220 are performed. Then, a process for sequentially adding stores that can be delivered on delivery days fixed to the key store to each new key store is performed, and a plurality of clusters each having a fixed delivery day are created.

  Note that the cluster creation unit 112, when calculating the number of restricted stores, does not set the number of stores for each day of the week for a store whose delivery day is not fixed yet, but is fixed to (the number of delivery days of the store / cluster). The number of days of the week that can be delivered to the store among the delivery days).

  Further, the cluster creation unit 112 can combine the created clusters. The cluster creation unit 112 combines clusters when the number of stores taken into each cluster is smaller than the limited number of stores. This process can be performed for each day of the week.

  The day of week allocation unit 114 sequentially fixes the delivery day of the week to each store when the delivery day of delivery to each store can be fixed in each of the above processes. For example, since the store with the store code “g” shown in FIG. 9 can have 2b and 2c as basic day patterns, the delivery day is not fixed yet. However, when the store with the store code “g” is taken into the cluster in which the delivery day not including Wednesday is fixed, the delivery day of the store is fixed to the basic day pattern 2b.

  The group creation unit 116 creates a group for each day of the week when a delivery day fixed to the cluster includes a plurality of days in each of the plurality of clusters, and the delivery day of the week is already fixed among the stores included in the cluster. Are classified into groups for each day of the week, and so that the difference in statistics calculated based on the stores included in the group for each day of the week is reduced, The stores included in the cluster are sequentially added to the group and classified into any day group.

FIG. 14 is a diagram schematically showing this processing.
Here, as an example, a case where the basic day pattern 2a is fixed as a delivery day will be described as an example. First, the stores whose delivery days are already fixed are connected by the MST for each day of the week to create the MST for each day of the week (FIG. 14A). Here, Monday MST and Thursday MST are created. Subsequently, the stores for which the delivery day of the week is not fixed are performed between the MSTs for each day of the week. Here, the sum of the branches of MST for each day of the week (movement time between stores) is made uniform. FIG. 14B is a diagram showing an example in which one store is taken in on Monday MST. The group creation unit 116 performs processing until all stores in the cluster are included in any group (MST).

  With the above processing, the day of the week for delivering the article to each store is determined.

  In the present embodiment, for each day of the week, the store classification unit 122 connects stores for which the day of the week is the delivery day of the week with MST, cuts the line segment, and divides it into a predetermined number. Stores included in each divided line segment correspond to stores included in each delivery course. That is, the delivery order to the store included in each divided line segment is determined later, and a delivery course is created. Next, the store classification unit 122 performs processing for equalizing the statistics of stores included in each MST divided into a predetermined number. Here, the statistic can be, for example, a delivery time. The predetermined number can be calculated based on a calculation result of a necessary number calculation processing unit 124 described later. That is, the required number calculation processing unit 124 calculates the number of delivery courses for each store in the area for each day of the week where the day of the week is fixed as the delivery day. The number of delivery courses calculated for each day of the week may be a predetermined number for dividing the MST for each day of the week.

The specific procedure is shown below. The store classification unit 122 performs the following processing for each day of the week.
(1) First, based on the shortest path travel time, a full-area MST including all stores of the day is created.
(2) The created line segment of the entire area MST is cut and divided into a predetermined number of MSTs. Moreover, the store classification | category part 122 divides | segments so that the statistics of each MST divided | segmented into the predetermined number may become as equal as possible. Here, the statistic can be a travel time required for sequentially visiting the stores included in each course from the base and returning to the base. Further, the statistic can be the total of the delivery amount of the articles to the stores included in each MST, and can also be considered together with the travel time and the total delivery amount.

(3) In the above (2), the whole area MST is divided so that the statistics of each course are equal. However, when an imbalance occurs, adjustment processing can be performed. This adjustment process can be performed, for example, by moving the store from the MST having the largest statistic to the smallest MST. At this time, if the store is moved directly between the two MSTs, there is a possibility that the stores located far away from each other may be fetched. Therefore, adjacent information is provided between the MSTs so that the store moves only between the adjacent MSTs. To. The adjacency information can be determined based on whether or not the adjoining area MST is adjacent.

  This process is schematically shown in FIG. FIG. 15 shows an example in which the entire area MST is divided into 1 to 6 MSTs. When each course is in such an arrangement, if one store is to be moved from MST “1” to course “4”, first the store is moved from MST “1” to “2”, and then MST “2”. It is possible to move one store from “3” to “3” and then move one store from “3” to “4”. As described above, the MST for each day of the week is created. Thereafter, in each MST, a delivery order to each store is determined, and a delivery course is created.

Next, processing of the necessary number calculation processing unit 124 will be described.
The required number calculation processing unit 124 calculates the necessary number of delivery courses when creating a plurality of delivery courses that deliver articles from a base to a plurality of stores (travel points) and return to the base within the area.

  Here, when calculating the number of delivery courses based on time, the required number of delivery courses P is calculated from the following equation. Note that the weighted store number M can be the number of stores weighted according to the number of deliveries per week for each store. For example, the number of weighted stores of a store whose delivery number per week is “3” is “3”.

Number of delivery courses P = weighted number of stores M / number of stores per delivery course N (Expression 1)

  Here, the number N of stores per delivery course is calculated by the following equation.

Number of stores per delivery course N = (Work time per delivery course A−Average value of round trip travel time between base and store D + Average value E of travel time between stores) / (Average value of work time at store) Average value of travel time between C + stores E) (Equation 2)

  FIG. 16 is a block diagram showing a configuration of the required number calculation processing unit 124 in the present embodiment. The required number calculation processing unit 124 includes a preprocessing unit 308, a first average value calculation unit 310, a second average value calculation unit 312, a store number calculation unit 314, and a delivery course number calculation unit 316.

  Here, the condition setting accepting unit 104 shown in FIG. 2 also accepts input of setting of various conditions necessary for calculating the number of delivery courses from the user. Of the setting inputs accepted by the condition setting accepting unit 104, information about the base is stored in the base information storage unit T2, information about the store is stored in the store information storage unit T3, and other information is stored in the condition setting storage unit T4. Remembered. In addition to the configuration shown in FIG. 5, the store information storage unit T3 can include a reciprocating travel time (minutes) between the base and the store and a store-store travel time (minutes).

  In the present embodiment, the store information storage unit T3 (delivery information storage unit 126) uses the travel time itself between the base and each store as information that can calculate the travel time between the base and each store. Is remembered.

  FIG. 17 is a diagram showing a part of the data structure of the inter-store travel time storage unit 328 included in the store information storage unit T3 in the present embodiment. The store-to-store travel time storage unit 328 stores store-to-store travel time between other stores for each store. That is, in this embodiment, the store information storage unit T3 (inter-store travel time storage unit 328) uses each store and another store as information that can calculate the travel time between each store and each other store. It remembers the actual travel time between them. Here, for example, in the store with the store code “a”, the travel time between the store with the store code “b” is “20 minutes” and the travel time with the store with the store code “c” is “30 minutes”. ”And the travel time between the store with the store code“ d ”is“ 20 minutes ”.

  FIG. 18 is a flowchart showing a procedure by which the required number calculation processing unit 124 calculates the number of delivery courses in the present embodiment. Hereinafter, description will be made with reference to FIGS. 1, 2, 16, and 17.

  First, the preprocessing unit 308 calculates the number of stores, the number of weighted stores M, and the average value C of work hours at the stores (step S400). In addition, the average value C of the work time in the store can be calculated in advance, and the average value C of the work time in the store can be stored in the store information storage unit T3.

  Subsequently, the first average value calculation unit 310 calculates the average value D of the reciprocating travel time between the base and the store (step S402). In this calculation, the first average value calculation unit 310 considers only the stores that are included in the area, and the stores that are within a predetermined ratio in the shortest round trip time with respect to the base, After the exclusion, the average value D of the round trip time between the base and the store is calculated. Thereby, as the average value D, an approximate value of the round trip time between the base and the store in each delivery course when actually creating the delivery course can be obtained, and the number of delivery courses can be calculated accurately. It is possible to obtain an average value D of the reciprocating travel time between the base and the store.

  In the present embodiment, the first average value calculation unit 310 weights each store according to the number of delivery days per week when detecting a predetermined ratio of stores to be considered. The procedure is shown below.

  First, the first average value calculation unit 310 refers to the store information storage unit T3, and for a store having a plurality of delivery days per week, the first average value calculation unit 310 reciprocates between a plurality of bases and stores for the delivery days. Prepare a set of hours.

  FIG. 19 is a diagram showing a set of reciprocation time between the base and the store prepared by the first average value calculation unit 310. For example, in the example shown in FIG. 5, the store with the store code “a” has six sets because the number of delivery days per week is six. Similarly, two sets of stores with the store code “g” are prepared because the number of delivery days per week is two. The first average value calculation unit 310 arranges the set of reciprocating time between the base and the store prepared in this way in the order of short reciprocating time between the base and the store.

  Next, the first average value calculation unit 310 detects a set boundary that is within a predetermined ratio with respect to the total number of sets in the order of short reciprocation time between the base and the store. Here, the predetermined ratio can be set to 80%, for example.

  For example, in the example shown in FIG. 19, it is assumed that the set boundary within 80% is between the store with the store code “i” and the store with the store code “j”. That is, in this case, the average value D of the round-trip travel time between the base and the store is calculated considering only the stores within 60 minutes of the round-trip travel time between the bases among the stores included in the area. Will be.

  The first average value calculation unit 310 divides the total value of the round-trip travel time between each set of bases and stores at a predetermined ratio to be considered by the number of sets at a predetermined ratio to be considered. The average value D of the reciprocation time between them is calculated.

  Returning to FIG. 18, the second average value calculation unit 312 calculates an average value E of travel time between stores (step S <b> 404). FIG. 20 is a flowchart illustrating a procedure in which the second average value calculation unit 312 calculates an average value E of travel time between stores.

  The 2nd average value calculation part 312 calculates the average value of the travel time between the stores of the said store first for every store. First, one of the stores in the area is selected as a target store (step S450). Subsequently, one of the deliverable days of the target store is selected (step S452). Next, among other stores that can be delivered on the day of the week selected in step S452, a predetermined number of stores are selected in the order of shorter travel time from the target store selected in step S450 (step S454). Here, the predetermined number can be set to “6”, for example. If the number of other stores that can be delivered on the day of the week selected in step S452 is less than the predetermined number, only the stores that can be delivered on the day of the week selected in step S452 are selected.

  Subsequently, the second average value calculation unit 312 calculates the total travel time of the target store selected in step S450 and the store selected in step S454, and is selected in step S454. The number of stores in the store is temporarily stored in the temporary storage unit T7 (step S456).

  Subsequently, it is determined whether or not the processing has been completed for all deliverable days of the target store (step S458). If there is a day of the week that has not yet been processed (NO in step S458), the process returns to step S452, and the same Repeat the process. On the other hand, if the processing is completed for all the deliverable days of the target store in step S458 (YES in step S458), the total travel time is divided by the number of stores, and the travel time between stores of the target store Is calculated and stored in the temporary storage unit T7 (step S460).

FIG. 21 is a diagram illustrating an example of an internal configuration of the temporary storage unit T7.
Here, a case where the target store is a store having the store code “a” will be described as an example.
In the store with the store code “a”, according to the example shown in FIG. 5, the deliverable day of the week is “Monday, Tuesday, Wednesday, Thursday, and Friday”. Therefore, the process described with reference to steps S452 to S456 is performed for each of the month, fire, water, tree, gold, and soil. As a result, it is assumed that the total travel time and the number of stores for each day of the week are as shown in FIG. The second average value calculation unit 312 calculates the total time of the total time of each day of the week and the total number of stores. As a result, the total time is 900 minutes and the number of stores is 35. The second average value calculation unit 312 can calculate the average value Q of travel time between stores of the store with the store code “a” Q = 900 minutes / 35 = 25.71 minutes in the process of step S460.

  Returning to FIG. 20, it is then determined whether or not processing has been completed for all stores in the area (step S462). If there is a store that has not yet been processed (YES in step S462), the processing returns to step S450. Repeat the same process. On the other hand, if the processing is completed for all stores in step S462 (NO in step S462), the total time of the average value Q of travel times between stores of all stores is divided by the number of stores in the area. The average value E of the travel time between the stores in this area is calculated (step S464).

  By such processing, the average value Q of the travel time between each store and other stores can be accurately calculated according to the characteristics of the stores included in this area. It is possible to calculate the average value E of the travel time between them with high accuracy.

  Returning to FIG. 18, the store number calculation unit 314 calculates the number N of stores per delivery course according to the above-described (Equation 2) (step S <b> 406). Next, the delivery course number calculation unit 316 calculates the delivery course number P according to (Equation 1) described above (step S408).

  According to store classification system 100 in the present embodiment, the required number of delivery courses can be accurately calculated according to the characteristics of the area.

  In the above, an example in which the number of delivery courses P is calculated for all stores in the area has been shown. However, when the delivery day of the week to each store is determined, the required number calculation processing unit 124 The number of delivery courses for each day of the week can be calculated for each day of the week with the store fixed as being the day of the delivery for each day of the week. As described above, with this result as a predetermined number, the store classification unit 122 cuts the MST for each day of the week and divides it into the predetermined number. This result can also be used as a reference for setting the number of tracks required for each day of the week.

(Second Embodiment)
In the present embodiment, when the day of week allocation processing unit 106 performs the process of determining the delivery day of each store, and when the required number calculation processing unit 124 calculates the number of delivery courses P, all stores in the area are displayed. It is different from the first embodiment in that it is not a target but only a store that actually sells products at a predetermined time among the stores.

  For example, if the goods to be delivered are products that require permission to sell, such as cigarettes, depending on the status of the permission application, when the products can be sold at each store, and when they will be closed Can be grasped. In the present embodiment, the case where each store is a store that sells tobacco will be described as an example. In the present embodiment, in addition to the configuration of the delivery information storage unit 126 shown in FIG. 5, the delivery information storage unit 126 includes a store type column, a related column, a start date column, an abolition date column, and the like. It further includes an application date column.

FIG. 22 is a diagram illustrating an example of the configuration of the delivery information storage unit 126 of the store information storage unit T3 in the present embodiment.
In addition to the configuration of the delivery information storage unit 126 shown in FIG. 5, the delivery information storage unit 126 further includes a store type column, a related column, a start date column, and an abolition date column. Also in the present embodiment, the delivery information storage unit 126 includes other fields similar to those shown in FIG. 5, but the description is omitted here.

  In the “start date” column and the “abolition date” column of the delivery information storage unit 126, the date when cigarette sales are started and the date when cigarette sales are stopped in each store, respectively. You can remember the day.

  In addition, for the sale of tobacco, a store that has received a sales permit (hereinafter referred to as “permitted store”) sells tobacco at a permitted store and a store at a destination (hereinafter referred to as “destination store”). May be sold. In addition, when selling at a branch store, there are a case where a cigarette as a product is delivered from a permitted store to a branch store and a case where tobacco is delivered directly to a branch store. When cigarettes are delivered from an authorized store to a destination store, it is necessary to add the amount of articles (cigarettes) to the authorized store to the amount of articles (cigarettes) delivered to the authorized store and deliver it to the authorized store. In the “store type” column, a code indicating whether each store is a permitted store or a branch store, and if the store is a branch store, whether the cigarette is delivered from the permitted store or directly to the store is stored. Is done. In the present embodiment, the symbol “1” indicates that the store is a permitted store, the symbol “2” indicates that the store is a branch store, and articles are delivered directly to the store. “3” indicates that the store is a branch store and goods are delivered from the authorized store to the store.

  In the “relation” column, the relationship between the permitted store and the branch store is stored. For example, a store with a store code “a” is a permitted store for stores with store codes “b” and “c”. For example, when considering the amount of articles to the store with the store code “a”, the amount of articles to the store with the store code “c” that is to be delivered from the authorized store among the outlet stores It is necessary to consider. On the other hand, for the store with the store code “b”, since the goods are delivered directly to this store, when considering the amount of goods to the store with the store code “a”, the store code “b” is sent to the store. There is no need to consider the amount of goods. Further, the application date of the store code “b” to the store depends on the application date of the store of the store code “a”.

  In the present embodiment, the condition setting accepting unit 104 can accept the input of the conditions of the target store for performing the process of determining the delivery day of the week to each store and the process of calculating the number of delivery courses. The condition setting reception unit 104 can receive, for example, input of the start date and the abolition date, and the store type of the target store. Here, for the store with the store type “3”, since it is not necessary to consider the delivery of the article to the store in the course creation, the condition setting reception unit 104 inputs the store type of the target store. , “1” and “2” can be received.

  In the present embodiment, the day-of-week allocation processing unit 106 and the required number calculation processing unit 124 of the store classification system 100 extract stores that meet the conditions received by the condition setting receiving unit 104 and target only the stores. The same processing as described in the first embodiment is performed.

  With the above processing, it is possible to assign a delivery day of the week to each store or calculate the required number of delivery courses with high accuracy according to the characteristics of the area when the delivery is actually performed.

  The number of delivery destination stores is enormous when considered at the national level. For example, taking an example of plotting a delivery destination store on a digital map, if the number of stores is enormous, the response of the process will be very poor. Therefore, in order to improve processing response, it is preferable not to read store data that is not required for processing. By the way, regarding the licensing business related to certain products such as cigarettes, it is considered that the replacement of stores that are started and abolished will become even more intense in the future. Therefore, it is not necessary to read the data of the abolished store, but it is not preferable to delete the abolished store as a physical record. In addition, by registering in advance as physical records for stores that will open in the future, it is possible to perform pre-simulation processing, so stores that have not yet started can also be registered as physical records. It is preferable to keep. According to the configuration of the present embodiment, the delivery information storage unit 126 is provided with a start date field and an abolition date field so that logical records can be read without reading all physical records. Can lead. Thereby, in the process of combining the basic technology (platform: digital map, for example) and the application, only the logical record can be sequentially read while considering the physical record, and the response can be improved.

  Each component of the store classification system 100 shown in FIG. 2 is not a hardware unit configuration but a functional unit block. Each component of the store classification system 100 is centered on an arbitrary computer CPU, memory, a program for realizing the components shown in the figure loaded in the memory, a storage unit such as a hard disk for storing the program, and a network connection interface. It is realized by any combination of hardware and software. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus.

  The present invention has been described based on the embodiments. The embodiments are exemplifications, and it will be understood by those skilled in the art that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are within the scope of the present invention. .

  The store classification system 100 according to the present embodiment can be used not only to classify stores included in each delivery course but also to calculate the number of delivery courses in order to actually create delivery courses. That is, when the store classification unit 122 classifies the stores to be included in each delivery course, the store classification system 100 refers to the delivery course number P calculated by (Equation 1) and the number of classifications is equal to or less than k−1. However, when the number of classifications = k, k that does not leave a store can be calculated as the optimal number of delivery courses.

  The procedure for classifying the stores to be included in each delivery course in order for the store classification unit 122 to create a plurality of delivery courses uses various procedures such as the method described in Patent Document 1 in addition to the process using MST. Can do.

  In the above embodiment, the store information storage unit T3 can calculate the travel time between the base and each store, and the travel time itself between the base and each store, and each store and others. The case where the travel time itself between each store and each of the other stores is stored as information that can calculate the travel time between each store has been described as an example. However, in another example, store information storage part T3 can memorize the position information on each base and each store as information which can calculate the movement time between each base and each store. In this case, the store classification system 100 can further include a travel time calculation unit that calculates a travel time between the base and each store based on the position information. In this case, the map information storage unit T1 can store road information such as the speed limit of the road between the stores and the shortest route. The travel time calculation unit can calculate the travel time between the base and each store based on position information and road information between the base and each store. The travel time calculation unit can also calculate the travel time between stores based on the location information and road information of each store. For example, when calculating the travel time between one store and another store, the travel time calculation unit first determines which road to use based on the location information and road information of these stores. To detect. Next, the travel time calculation unit calculates the travel time between stores based on the speed limit of the road to be used and the travel distance.

  In the above embodiment, the case where the number of key stores selected by the key store selection unit 110 is the number of required delivery courses calculated by the required number calculation processing unit 124 has been described as an example. However, as another example, the key store selection unit 110 can select a number of key stores corresponding to the number of tracks set in advance in the area, for example.

Such setting of track information can be stored in the condition setting storage unit T4. FIG. 23 is a diagram illustrating a part of the data structure of the track information storage unit 134 stored in the condition setting storage unit T4.
The track information storage unit 134 stores, for each day of the week, the size and number of available tracks, the delivery time of each track, and the like. Here, for example, 4 ton trucks are set from Monday to Friday, the delivery time of each truck is 6 hours, and on Saturday, 2 ton trucks are set and the delivery time of each truck is 6 hours. Here, the maximum delivery time of one course can be set to, for example, 6 hours in correspondence with the deliverable time of each track shown in FIG.

  Although not described here, the track information storage unit 134 can also store information such as the number of track workers and the type of track. For example, when the store is in an area with poor security, the worker can perform delivery using a truck set as “taken” by a plurality of persons. Further, when it is necessary to pass through a road with a height restriction or to park in a place with a height restriction, delivery can be performed by a truck that can freely adjust the vehicle height. Such information can also be stored in association with each store of the delivery information storage unit 126 shown in FIG.

  Based on the delivery day fixed to the store selected as the key store and the track information stored in the track information storage unit 134, the key store selection unit 110 prevents the number of courses from exceeding the number of tracks on that day of the week. To control. For example, when a store having the store code “a” in FIG. 9 is selected as a key store, a cluster created from this key store as a starting point includes one delivery course from Monday to Saturday. . Therefore, when the selection of the store code “a” is selected as the key store, one month to Saturday delivery course has already been created. On the other hand, in the example shown in FIG. 23, the number of delivery courses is 4 for Monday through Friday, and the number of delivery courses is 2 on Saturday. Therefore, after the store with the store code “a” is selected, it is possible to create three delivery courses for Monday through Friday and one delivery course on Saturday. If the key stores are sequentially selected and four stores including Monday as a delivery day are selected as key stores, stores including Monday as a delivery day cannot be selected any more.

It is a flowchart which shows an example of the process sequence of the shop classification method in embodiment of this invention. It is a block diagram showing an example of composition of a store classification system in an embodiment of the invention. It is a flowchart which shows in detail the process sequence of the day-of-week allocation process part corresponding to step S20 of FIG. It is a block diagram which shows the structure of the shop information storage part in embodiment of this invention. It is a figure which shows a part of data structure of the delivery information storage part in embodiment of this invention. It is a block diagram which shows an example of a structure of the condition setting memory | storage part in embodiment of this invention. It is a figure which shows a part of data structure of the pattern memory | storage part in embodiment of this invention. It is a figure which shows a part of data structure of the day fixed rule memory | storage part in embodiment of this invention. It is a figure which shows a part of data structure of the day allocation information storage part in embodiment of this invention. It is a flowchart which shows the procedure of step S106 of FIG. It is a flowchart which shows the procedure of step S210 of FIG. It is a figure which shows typically the process of step S210 of FIG. It is a figure which shows typically the process of step S210 of FIG. It is a figure which shows typically the process which a group preparation part produces the group for every day of the week in a cluster. It is a figure which shows typically each MST which the shop classification part produced. It is a block diagram which shows an example of a structure of the required number calculation process part in embodiment of this invention. It is a figure which shows a part of data structure of the store travel time memory | storage part. It is a flowchart which shows the process sequence of the store classification system in embodiment of this invention. It is a figure which shows the set of the reciprocating movement time between the base and store which were prepared by the 1st average value calculation part. It is a flowchart which shows the procedure in which the 2nd average value calculation part calculates the average value E of the travel time between stores. It is a figure which shows an example of an internal structure of a temporary memory part. It is a figure which shows an example of a structure of the delivery information storage part of the store information storage part in embodiment of this invention. It is a figure which shows a part of data structure of the track information storage part of the track information storage part contained in a condition setting storage part.

Explanation of symbols

100 store classification system 102 display processing unit 104 condition setting reception unit 106 day allocation processing unit 108 pattern allocation unit 110 key store selection unit 112 cluster creation unit 114 day allocation unit 116 group creation unit 118 adjustment processing unit 120 calculation unit 122 store classification unit 124 Required number calculation processing unit 126 Delivery information storage unit 128 Day allocation information storage unit 130 Pattern storage unit 132 Day fixed rule storage unit 134 Track information storage unit 200 Base 202 Store 204 MST
206 spare cluster 208 spare cluster 210 spare cluster 212 spare cluster 308 preprocessing unit 310 first average value calculating unit 312 second average value calculating unit 314 store number calculating unit 316 delivery course number calculating unit 328 inter-store travel time storage unit T1 Map information storage unit T2 Base information storage unit T3 Store information storage unit T4 Condition setting storage unit T5 Course information storage unit T6 Logic storage unit T7 Temporary storage unit

Claims (15)

In order to create a plurality of delivery courses for delivering articles to a plurality of circulation points included in a predetermined area, the system classifies the circulation points included in each of the delivery courses,
A traveling point information storage unit for storing the number of days of delivery per week and the day of delivery available to the traveling point, and the position information of the traveling point, for each traveling point;
A day-of-week assignment processing unit that assigns a delivery day for delivering the article to the circulation point for each of the circulation points;
For each day of the week, a traveling point classifying unit that classifies the traveling points to be included in each of the delivery courses in order to create a delivery course that includes each of the traveling points assigned by the day of the week allocation processing unit as the delivery day of the week. When,
Including
The day of week allocation processing unit
A day of week allocation unit for fixing a delivery day of the week to the patrol point where the delivery day of the week can be fixed based on the number of delivery days and the day of delivery;
A key tour point selection unit that selects a plurality of tour points as key tour points so as to be distributed within the area based on position information of each tour point from among the tour points whose delivery day has already been fixed. When,
A plurality of clusters including each of the key circulation points, wherein the delivery day of the week fixed to the key circulation point is fixed as a delivery day of the week, and the circulation points that can be delivered on each of the delivery days are sequentially assigned to each of the clusters. A cluster creation unit that performs processing to add and classifies the traveling points included in the area into any of the clusters;
In each of the plurality of clusters, when the delivery days fixed to the cluster include a plurality of days, a group for each day is created, and the delivery day is already fixed among the tour points included in the cluster. The distribution day of the week is not yet fixed so that the difference in the statistics calculated based on the tour points included in the group of the day of week is reduced. A group creation unit that sequentially adds the traveling points included in the cluster to any day group,
Including
The day of week allocation unit is configured to fix the tour day in which the delivery day of the week can be fixed based on the delivery day of the week, the number of delivery days of each tour point, and the day of delivery available for each of the plurality of clusters created by the cluster creation unit. A traveling point classification system for fixing a day of a delivery to a point and fixing a day of the group as a day of delivery to the traveling point included in the group of each day of the week created by the group creation unit. In the traveling point classification system according to claim 1,
Based on the location information of each of the tour points, further includes a calculation unit that calculates a travel time between the tour points included in the group for each day of the week,
The group creation unit is configured such that the delivery day is not yet fixed as the statistic so that the difference in the total travel time between the traveling points included in the group for each day of the week is reduced. A traveling point classification system that adds to the group of days of the week. In the traveling point classification system according to claim 1 or 2,
The day of week allocation unit fixes the day of delivery to the tour point as the day of delivery of the tour point when the number of delivery days of the tour point is equal to the number of days of the deliverable to the tour point. Classification system. In the traveling point classification system according to any one of claims 1 to 3,
A pattern storage unit for storing a basic day of the week pattern in which a delivery day is specified for each delivery day per week;
The day of the week allocating unit fixes a day of the week specified in the basic day of week pattern as a delivery day of the tour point that can be delivered on the day of the week specified in the basic day of the week pattern. In the traveling point classification system according to claim 4,
The basic day of the week as a delivery day of the circulation point that can be delivered on the day of the week specified by the basic day pattern before the selection of the key circulation point by the key circulation point selection unit for each delivery day per week A weekday fixed rule storage unit for storing a rule as to whether or not to fix the day of the week specified in the pattern;
The day allocation unit is a tour point classification system that fixes a delivery day of each tour point according to the rule of the day fix rule storage unit. In the traveling point classification system according to any one of claims 1 to 5,
The key circulation point selection unit
A first key circulation point candidate selection step of selecting the circulation point whose delivery day of the week is already fixed in the area as a key circulation point candidate;
Classifying the plurality of circulation points into a plurality of spare clusters so as to be distributed in the area, based on position information of each of the circulation points;
In the spare cluster, based on the delivery days and the deliverable days of the week of the round points included in the spare cluster, when the round day in which the delivery day is already fixed is not included in the spare cluster, The delivery day with the highest delivery necessity is fixed as the delivery day of the spare cluster, and one of the tour points that can be delivered on the delivery day is selected, and the delivery day of the visit point is selected as the delivery day of the tour point. A second key tour point candidate selection step of fixing the delivery day fixed to the standby cluster and selecting the tour point as a key tour point candidate;
A cyclic point classification system for selecting the key cyclic point from the key cyclic point candidates selected in the first key cyclic point candidate selection step and the second key cyclic point candidate selection step. In the traveling point classification system according to claim 6,
The key circulation point selection unit includes a delivery fixed to the circulation point from the key circulation point candidates selected in the first key circulation point candidate selection step and the second key circulation point candidate selection step. A traveling point classification system that preferentially selects those having a large number of days of the week as the key traveling points. A traveling point classification method for classifying the traveling points to be included in each of the delivery courses in order to create a plurality of delivery courses for delivering articles to a plurality of traveling points included in a predetermined area,
For each traveling point, reading data from a traveling point information storage unit that stores the number of delivery days per week to the traveling point and the day that can be delivered, and location information of the traveling point;
Assigning a delivery day for delivering the article to the circulation point for each of the circulation points;
For each day of the week, in the step of assigning the delivery day of the week, the step of categorizing the circulation points to be included in each delivery course in order to create a delivery course that includes each of the tour points assigned as the delivery day of the week;
Including
The step of assigning the delivery day includes
Fixing the delivery day of the week to the patrol point where the delivery day can be fixed based on the delivery days and the deliverable day of the week for each round point;
Selecting a plurality of the tour points as key tour points so as to be distributed in the area based on the location information of the visit points from the tour points where the delivery day is already fixed;
A plurality of clusters including each of the key circulation points, wherein the delivery day of the week fixed to the key circulation point is fixed as a delivery day of the week, and the circulation points that can be delivered to each of the delivery days are sequentially assigned to each of the clusters. Performing a process of adding, and classifying the cyclic points included in the area into any of the clusters;
In each of the plurality of clusters, when the delivery days fixed to the cluster include a plurality of days, a group for each day is created, and the delivery day is already fixed among the tour points included in the cluster. The distribution day of the week is not yet fixed so that the difference in the statistics calculated based on the tour points included in the group of the day of week is reduced. Sequentially classifying the tour points included in the cluster into any of the day groups,
Including
The step of fixing a delivery day of the week to the round point includes the day of delivery fixed to each of the plurality of clusters created in the step of classifying into the cluster, the number of days of delivery of the round point, and the day of delivery available. Based on the day of the group, the day of delivery of the group is fixed as the day of delivery to the rounding point included in the group of the day of week created in the step of classifying the group into the group. Cyclic point classification method to fix. In the traveling point classification method according to claim 8,
Further comprising calculating a travel time between the tour points included in the group for each day of the week based on the location information of each tour point;
In the step of classifying into the day of the week group, the delivery day of the week is not yet fixed as the statistic so that the difference in the total travel time between the traveling points included in the group for each day of the week is reduced. A traveling point classification method for sequentially adding things to the group of each day of the week. In the traveling point classification method according to claim 8 or 9,
The step of fixing the delivery day of the week to the tour point is, when the number of delivery days of the visit point and the number of deliverable days of the visit to the visit point are equal, A method of classifying traveling points that is fixed as a delivery day. The traveling point classification method according to any one of claims 8 to 10,
Further including reading data from a pattern storage unit that stores a basic day of the week pattern in which a delivery day of the week is specified for each delivery day per week;
The step of fixing a delivery day of the week to the tour point includes a tour point classification method of fixing a day of the week specified by the basic day pattern as a delivery day of the tour point that can be delivered on the day of the week specified by the basic day pattern. . The traveling point classification method according to claim 11,
Before the selection of the key tour point in the step of selecting as the key tour point for each delivery day per week, as the delivery day of the tour point that can be delivered on the day specified by the basic day pattern, Further comprising the step of reading data from a day of the week fixed rule storage that stores a rule as to whether or not to fix the day of the week specified in the basic day of the week pattern.
The step of fixing the delivery day of the week to the circulation point is a circulation point classification method of fixing a delivery day of the circulation point according to the rule of the day of the week fixed rule storage unit. The traveling point classification method according to any one of claims 8 to 12,
The step of selecting as the key circulation point is:
A first key circulation point candidate selection step of selecting the circulation point whose delivery day of the week is already fixed in the area as a key circulation point candidate;
Classifying the plurality of circulation points into a plurality of spare clusters so as to be distributed in the area, based on position information of each of the circulation points;
In the spare cluster, based on the delivery days and the deliverable days of the week of the round points included in the spare cluster, when the round day in which the delivery day is already fixed is not included in the spare cluster, The delivery day with the highest delivery necessity is fixed as the delivery day of the spare cluster, and one of the tour points that can be delivered on the delivery day is selected, and the delivery day of the visit point is selected as the delivery day of the tour point. A second key tour point candidate selection step of fixing the delivery day fixed to the standby cluster and selecting the tour point as a key tour point candidate;
A cyclic point classification method for selecting the key cyclic point from the key cyclic point candidates selected in the first key cyclic point candidate selection step and the second key cyclic point candidate selection step. In the traveling point classification method according to claim 13,
The step of selecting as the key circulation point is fixed to the circulation point from the key circulation point candidates selected in the first key circulation point candidate selection step and the second key circulation point candidate selection step. A circulating point classification method for preferentially selecting a key having a large number of days of delivery as the key circulating point. A program for classifying traveling points to be included in each delivery course in order to create a plurality of delivery courses for delivering articles to a plurality of traveling points included in a predetermined area,
Computer
For each traveling point, a traveling point information storage means for storing the number of delivery days per week to the traveling point and the day of the week that can be delivered, and the position information of the traveling point;
Day-of-week assignment processing means for assigning a delivery day of the week for delivering the article to the circulation point for each of the circulation points;
For each day of the week, a traveling point classification means for classifying the traveling points to be included in each delivery course in order to create a delivery course that includes each of the traveling points assigned by the day of the week allocation processing means as the delivery day of the week;
Function as
The day of week allocation processing means includes:
A day-of-week assignment means for fixing a delivery day of the week to the circulation point where the delivery day of the week can be fixed based on the number of delivery days and the day of delivery available;
Key traveling point selection means for selecting a plurality of traveling points as key traveling points so as to be distributed within the area based on the position information of each traveling point from among the traveling points whose delivery day is already fixed. ,
A plurality of clusters including each of the key circulation points, wherein the delivery day of the week fixed to the key circulation point is fixed as a delivery day of the week, and the circulation points that can be delivered to each of the delivery days are sequentially assigned to each of the clusters. Cluster creation means for performing a process of adding, and classifying the cyclic points included in the area into any of the clusters,
In each of the plurality of clusters, when the delivery days fixed to the cluster include a plurality of days, a group for each day is created, and the delivery day is already fixed among the tour points included in the cluster. The distribution day of the week is not yet fixed so that the difference in the statistics calculated based on the tour points included in the group of the day of week is reduced. Group creation means for sequentially adding to the group of each day of the week and classifying the tour points included in the cluster into any of the groups of the day of the week;
Including
The day of week allocation unit is configured to fix the tour day in which the delivery day of the week can be fixed based on the delivery day of the week, the number of delivery days of each tour point, and the day of delivery available for each of the plurality of clusters created by the cluster creation unit. A program for fixing a day of the week for delivery to a point and fixing a day of the group as a day of delivery to the traveling point included in the group for each day of the week created by the group creation unit.

Images