JP2004238129A - Delivery plan planning method and its device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a delivery plan planning method capable of performing delivery planning of eliminating combination where distance between delivery destinations is long and delivering cargo to a plurality of delivery destinations at low density and in a large region, and to provide its device. <P>SOLUTION: The delivery plan planning device comprises an input device for inputting master data information related to the information on each order and the delivery plan planning, a processing device for creating a delivery plan where a plurality of delivery vehicles deliver cargo to the plurality of delivery destinations, an output device for outputting the created delivery plan, and an order grouping means for dividing the order and creating groups based on distance data master between the delivery destinations, and executes planning processing of the delivery plan by divided order group. In the delivery plan planning method, based on distance data master between the delivery destinations, combinations x1-x8 between the delivery destinations between which distance is not longer than a previously designated distance are extracted, the delivery destinations are grouped (Gr-1, 3, 7), and an order group is created based on the grouping. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
[Field of the Invention]
The present invention relates to a delivery plan formulation method and apparatus capable of departure from a distribution base or warehouse in a distribution system and creating a delivery plan for distributing a load even to a plurality of delivery destinations having a wide area and low density.
[0002]
[Prior art]
The problem of the distribution plan in the distribution system is that a plurality of vehicles depart from a common distribution base and distribute the cargo to a plurality of destinations, so as to find one that minimizes the total mileage. Each vehicle has restrictions such as upper limit of the loading capacity, vehicle specification such as a vehicle equipped with a UNIC, a vehicle having a specified possible loading weight, and unloading time specification.
[0003]
For example, when a vehicle is specified, a large vehicle cannot be delivered because the receiving site of the load is narrow or the road to the site is narrow, etc. There is a designation. In addition, when specifying the unloading time, there is no luggage storage at the site, and the arrival time of the luggage is specified according to the construction progress, or a truck arriving at the site earlier than the specified time is parked on the street and traffic congestion occurs This is performed in order to prevent the occurrence of, for example.
[0004]
The step of creating a delivery plan created under such restrictions is usually divided into two stages. One is a process of distributing the delivery destination package to each vehicle. The other is a process of optimizing a route in which the vehicle circulates the luggage loaded in each vehicle. Normally, these two processes are performed so that the loading ratio is increased so as to reduce the number of vehicles used while observing delivery restrictions, and furthermore, the mileage when vehicles travel around the delivery destination is shortened. The delivery order has been determined.
[0005]
Patent Literature 1 describes a method in which a delivery destination is represented in advance by a two-dimensional plane of a map, a fan-shaped partial space is created around a delivery base, and a vehicle allocation plan is implemented for each area of the partial space. . In the method, a plurality of delivery planning processes of algorithms having different planning strategies, such as a distance priority method, a direction division method, and the like, are executed in a single processing device in a time-division manner or in parallel by a plurality of processing devices. Among them, the processing result that minimizes the total traveling distance is selected and output from the output device.
[0006]
In the direction division (method), the space (delivery target area) is divided by a straight line passing through several logistics bases, and for each fan-shaped subspace generated, the delivery destination package located in that direction A number of seeds that are proportional to the sum of the quantities are allocated to the area.
[0007]
[Patent Document 1]
JP-A-2001-188884
[Problems to be solved by the invention]
However, in the conventional delivery planning method, the process of allocating the package of the delivery destination to each vehicle and the process of optimizing the route in which the vehicle circulates the package loaded in each vehicle are performed alternately. Finding the final optimal solution.
[0009]
When the delivery destination of the package of each order is nationwide, if these two processes are handled by handling all the packages of the order collectively, for example, when the delivery destination is nationwide, the package in the Kanto and Kansai regions Is not efficient to deliver on the same truck. Therefore, in the final delivery plan, a plan is made so that luggage in the same area is loaded on one truck as much as possible.
[0010]
At this time, if all luggage is processed at once and search is performed, an inefficient combination of luggage to be loaded on one truck will occur, processing time for planning will increase, and the number of searches will increase. May not be able to reach the optimal solution.
[0011]
Further, as in the technique described in Patent Document 1, the delivery destination is represented in advance by a two-dimensional plane of a map, a fan-shaped partial space is created around the delivery base, and a vehicle allocation plan is implemented for each area of the partial space. In the case of the method, if there are delivery destinations near the boundary of each adjacent fan type, these delivery destinations will be dispatched in each direction, so that no luggage will be loaded on the same vehicle There is a problem that is dispatched to.
[0012]
Alternatively, if the order is divided into prefectures, for example, in order to reduce unnecessary search, the crossing of destinations facing the prefectural border will be excluded from the beginning, and it will not be the target of delivery planning. You may miss a chance.
[0013]
The present invention eliminates combinations in which distribution destinations are long distances before departure processing, departs from distribution bases or warehouses in a distribution system, and delivers cargo to a plurality of distribution destinations with a large area and low density. It is an object of the present invention to provide a delivery plan drafting method and an apparatus thereof that enable a delivery plan to be made.
[0014]
[Means for Solving the Problems]
The above problem is solved by the following invention. According to the invention, an input device for inputting information of each order and input data of master data relating to delivery planning, and a plurality of delivery vehicles leave a distribution base and deliver a load to a plurality of delivery destinations based on the input information. A processing device for creating a delivery plan, an output device for outputting the created delivery plan, and order grouping means for dividing the order based on a master of distance data between delivery destinations and creating a group, A delivery plan drafting apparatus characterized in that a delivery plan planning process is executed for each divided order group.
[0015]
Further, in the present invention, the information of each order includes at least one of a delivery destination, a package, a weight, an unloading designation time, and a delivery vehicle designation of each order, and the master data related to the delivery planning is at least from the distribution base. The processing device for creating a delivery plan including one or more of a distance to a delivery destination, distance data between delivery destinations, and a speed pattern creates a delivery plan that minimizes the total mileage on a delivery route. May be used as a delivery planning device.
[0016]
Further, using the delivery planning device of these inventions, the destinations of each order are extracted based on the distance data master between the destinations, and the combinations between the destinations less than a predetermined distance are extracted to group the destinations. The invention can also be an invention of a delivery planning method, wherein the order group is created based on the grouped delivery destinations.
[0017]
In these inventions, first, the order is divided according to the delivery destination, so that the process of drafting the delivery plan can be executed for each of the divided order groups. As a result, the amount of calculation required for the search can be significantly reduced. For example, when long-distance delivery destinations such as Aomori and Fukuoka are included in the order for which a delivery plan is made, it is practically impossible to deliver these orders by the same vehicle. However, if an order including these delivery destinations is collectively drafted for a delivery plan, a combination of these may occur in the calculation process in the course of drafting, resulting in unnecessary search. .
[0018]
Here, the master of the distance data between the delivery destinations is originally a master necessary for determining the delivery time and the delivery order, but can also be used as a master when dividing the delivery destinations, This can be achieved without preparing a new master for this purpose.
[0019]
Based on the distance data master between destinations, extract the destinations of each order, and extract combinations between destinations that are less than or equal to the specified distance, and group the extracted combinations that have the same destination. As a result, the search space can be narrowed, and the search can be speeded up.
[0020]
For example, when three orders of 300 orders are loaded on one truck, when the order is divided into ten groups, as shown below, only one hundredth of the combination without division is required.
(1) Combination without order division:
(300 × 299 × 298) / (3 × 2 × 1) = 4,455,100 ways (2) Combination in the case of order division:
(30 × 29 × 28) / (3 × 2 × 1) × 10 = 40,600 ways In this way, by grouping combinations having the same delivery destination and dividing the order, The order can be divided in advance such that there is no short-distance delivery destination order other than the group of. As a result, since a delivery plan is created by a narrowed search, the speed can be increased.
[0021]
In addition, as the distance data, when there is traffic congestion such as in an urban area, the travel time between delivery destinations becomes longer and the travel range of vehicles becomes narrower in a city where there is traffic congestion than in a rural area. Therefore, if the target search area is narrowed so that the inter-delivery distance in an urban area is shorter than in a rural area, the search time is further shortened.
[0022]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, the present invention will be described in detail based on the illustrated embodiments. FIG. 1 shows an embodiment of the configuration of an optimal delivery planning system according to the present invention.
[0023]
The order information necessary for the vehicle allocation plan is input from the operation terminal of the factory / sales office and transmitted to the server. The order information includes attached information such as the delivery destination of the order, the weight and quantity of the item, the delivery date and time, and the presence or absence of the designated vehicle. A dispatch plan formulation operation is executed from the dispatch plan formulation section terminal based on the order information DB and the master DB of the order information collected in the server.
[0024]
FIG. 2 shows the procedure of vehicle allocation planning. As shown in FIG. 2, first, in S1, the vehicle allocation plan target order data is fetched. Next, in S2, based on the order information, master data of a target portion of the vehicle allocation plan is fetched from a master DB (delivery destination master, destination distance, data between destinations, etc.). Before executing the automatic dispatching plan, if there is a correction in the plan conditions such as stacking, vehicle designation, and delivery date and time in S3, the correction processing is performed by setting the dispatching plan execution condition.
[0025]
In a state where the conditions necessary for the automatic vehicle allocation plan are prepared, the vehicle allocation plan of S4 is executed. First, a distance calculation (step 1) necessary for dividing the delivery area is executed. The flowchart is shown in FIG. For example, in step 1, there are ten destination cities A to J as shown in FIG. 5, and the distance between the cities is calculated. As a result, a combination in which the inter-city distance (FIG. 3d (i, j)) based on the actual road distance on the map is equal to or less than the predetermined distance (FIG. 3d # param) is the cell (i) in FIG. , J) are stored as distances x1 to x8.
[0026]
For the distance between cities, for example, if the customer is a fixed customer, the distance data between the customers is used as the master. If the customer constantly changes, the municipalities of the delivery address are set as the representative values and managed by the municipal JIS code of the whole country.
[0027]
The order grouping means performs delivery area division, that is, grouping of delivery destinations (city, fixed customer) (step 2), based on the inter-city distance table shown in FIG. The flowchart is shown in FIG. Here, the presence or absence of the inter-city distance is determined for the cell at the i-th row and the j-th column of the table, and the inter-city distance sharing the row or column among the cells in which the inter-city distances x1 to x8 are described. Group.
[0028]
In the table shown in FIG. 5, there are ten destinations from A to J, and A to J are assigned to Gr-1 to Gr-10 (FIG. 4 group # id (i) = i). The grouping process searches for a cell that is not blank (null) from the upper row to the lower row of each row in order from the left column in FIG. First, when j (column) = 2, i (row) = 1 (column B, row A), the cells (cell (i, j) in FIG. 4) are not blank, and the intercity distance x1 exists. A and B belong to the same group. At this time, the delivery destination B is aligned with the group ID of the delivery destination A (FIG. 4, group # id (j) = group # id (i)). Therefore, the set of each group at this stage is as shown in equation (1).
[0029]
Gr-1 = {A, B}, Gr-3 = {C}, Gr-4 = {D},
Gr-5 = {E}, Gr-6 = {F}, Gr-7 = {G},
Gr-8 = {H}, Gr-9 = {I}, Gr-10 = {J} (1)
Next, when j = 3 (column C), no grouping is performed because there is no description of the distances x1 to x8 between cities. When j = 4 (column D), since there is an inter-city distance x2 at i = 3 (row C), these are grouped. In this case, the delivery destination D is aligned with the group ID of the delivery destination C. Therefore, the set of each group at this stage is represented by the expression (2).
[0030]
Gr-1 = {A, B}, Gr-3 = {C, D}, Gr-5 = {E},
Gr-6 = {F}, Gr-7 = {G}, Gr-8 = {H},
Gr-9 = {I}, Gr-10 = {J} (2)
Furthermore, when j = 5 (column E), since i = 1 and 2 have inter-city distances x3 and x4, they are grouped with the delivery destinations A and B, so that the delivery destination E is Gr-1. Therefore, the set of each group is given by equation (3).
[0031]
Gr-1 = {A, B, E}, Gr-3 = {C, D},
Gr-6 = {F}, Gr-7 = {G}, Gr-8 = {H},
Gr-9 = {I}, Gr-10 = {J} (3)
If the above processing is continued up to j = 10 (J column), the set of groups becomes the equation (4).
[0032]
Gr-1 = {A, B, E}, Gr-3 = {C, D, F, J},
Gr-7 = {G, H, I} (4)
By the above method, if the number of cities is n, the calculation of the distance between cities is n (n-1) / 2 times, the judgment of the presence or absence of the group ID is n (n-1) / 2 times, and the assignment of the group ID is The number is n, and the total amount of calculation is n ² times.
[0033]
In this way, by the process of step 2 (FIG. 4), 10 cities can be divided into three groups from the state of FIG. 5A as shown in FIG. 5B. As a result, Gr-1 becomes A, B, E, Gr-3 becomes C, D, F, J, and Gr-7 becomes G, H, I.
[0034]
As a result, the delivery area is divided into three, and a vehicle allocation plan can be formulated for each group. The orders are also grouped based on the division of the delivery area into groups. For each group, taking into account the constraints of the order information (delivery time specification, vehicle type specification, etc.) and the constraints in planning (loading rate, order consolidation of the same shipper, etc.), the loading rate or the number of vehicles As a search problem using the mileage as an evaluation function, a solution of a vehicle allocation plan is obtained. As a solution search method, a genetic algorithm, a generalized assignment problem, or the like is applied.
[0035]
The vehicle allocation plan results calculated for each group can all be confirmed from the vehicle allocation plan formulation section terminal in FIG. 1 and an interface is also provided that allows the contents of the plan results to be modified. After confirming and correcting the contents of the dispatch plan (S5 in FIG. 2) at the dispatch planning unit terminal in FIG. 1, if there is no problem (S6 in FIG. 2), the finalization and transmission (S7 in FIG. 2) are operated from the terminal. I do.
[0036]
【The invention's effect】
According to the present invention, the order is divided according to the delivery destination, so that the process of planning the delivery plan can be executed for each of the divided order groups. Furthermore, by providing delivery destination grouping means for extracting a combination between delivery destinations of a predetermined distance or less and creating a group of extracted orders based on the combination, it is possible to greatly reduce the amount of calculation required for the search. As a result, it becomes possible to carry out a delivery plan for delivering a load to a plurality of delivery destinations having a wide area and low density.
[Brief description of the drawings]
FIG. 1 is a configuration diagram showing an embodiment of a delivery planning system according to the present invention.
FIG. 2 is a flowchart showing a procedure for formulating a plan according to the present invention.
FIG. 3 is a flowchart for calculating a combination of distances equal to or less than a distance specified between delivery destinations.
FIG. 4 is a flowchart illustrating a distribution destination area division algorithm.
FIG. 5 is a diagram illustrating an example of division of cities in a delivery destination area. (A) State before division (b) State after division

Claims (3)

An input device for inputting the information of each order and the input information of the master data relating to the planning of the delivery plan, and a delivery plan in which a plurality of delivery vehicles leave the distribution base and deliver the cargo to a plurality of delivery destinations based on the input information. A processing device for creating the order, an output device for outputting the created delivery plan, and order grouping means for creating a group by dividing the order based on a master of distance data between delivery destinations. A delivery plan drafting apparatus for executing a delivery plan drafting process for each group. The information of each order includes at least one of the delivery destination, package, weight, unloading designation time, and delivery vehicle designation of each order, and the master data related to the delivery planning is at least the distance from the distribution base to the delivery destination. And a processing device for creating a delivery plan including one or more of distance data between destinations and a speed pattern, wherein the processing device creates a delivery plan that minimizes the total mileage on the delivery route. The described delivery planning device. Using the delivery planning device according to claim 1 or 2, a combination of delivery destinations of a predetermined distance or less is extracted and delivered based on a distance data master between delivery destinations for each order. A method of making a delivery plan, comprising grouping destinations and creating the order group based on the grouped destinations.

Images