KR20010075961A - Method of vehicle assignment for logistic workflow - Google Patents

Abstract

PURPOSE: A vehicle allocating method for a logistics workflow and a record medium reading an allocating program through a computer are provided to embody a matching system saving a logistics time and cost by efficiently connecting demand and supply. CONSTITUTION: To allocate vehicle for a logistics workflow, an allocating system models a single garage or plural garages, a sort of a cargo or plural sorts of cargoes, a sort of a vehicle or plural sorts of vehicles and plural demands, in a base model. Then, the system outputs the shortest and most efficient delivery passage for the base model. After that, the system allocates vehicles according to vehicle capacity per the shortest and most efficient delivery passage.

Description

Method of vehicle assignment for logistic workflow

The present invention is a logistics that can implement a matching service that can save logistics costs and time through the efficient connection between the transport demand of the cargo generated in each region in the logistics information system and the transporter owning the vehicle as a means of transportation A vehicle dispatch method for a workflow and a computer-readable recording medium having recorded thereon a program for realizing the method.

The vehicle route problem is a generalized problem that adds various constraints such as vehicle capacity and maximum travel time to the multiple salesman problem. That is, the present invention deals with all problems of determining a vehicle route for performing a service to a customer using a vehicle. Application examples of such vehicle route problems include product delivery from dealerships, newspaper delivery, railroad operation, public transportation vehicle operation, airplane operation, patrol car operation, school bus operation, garbage truck operation, truck operation in other industries, etc. Very diverse The purpose of these problems is to minimize costs (ie, distribution costs, purchase of vehicles and warehouses, etc.) or to improve services (ie, shortening distribution times, etc.).

Characteristics of problems classified by "Bodin" and "Schrage" include vehicle base index, number of vehicles, type of vehicle, demand type, demand location, network direction, vehicle loading capacity, maximum driving time, arrival time to a specific point. Cost, objective function, etc.

The basic vehicle route problem is to find a vehicle route for minimizing the total travel distance between several vehicles having the same loading capacity, starting from the main store, delivering the product to each point where the demand is known and returning to the main store. Combining the above-mentioned characteristics in this problem, the salesman problem (TSP), the multiple traveling salesman problem, the multi-depot vehicle routing problem, the number of vehicles and the vehicle procurement Fleet Size and Mix Problems, Stochastic Vehicle Routing Problems, and Postman Problems are derived. In addition, there are problems using the branch limit method, the problem of calculating the appropriate number of vehicles while minimizing the total travel distance, and the asymmetric VRP problem addressed by "Laporte", and so on. There are problems such as finding an optimal solution based on a mathematical model.

Therefore, a method of implementing a matching service that can save logistics costs and time through an efficient connection between the transport demand of the cargo generated in each region and the transporter owning the vehicle, which is a transport vehicle, is essential in the logistics information system.

The present invention devised to meet the demand as described above, the logistics information system can save the logistics cost and time through the efficient connection between the transport demand of the cargo generated in each region and the transporter owning the vehicle as a means of transportation An object of the present invention is to provide a vehicle dispatch method for a logistics workflow for implementing a matching service and a computer-readable recording medium on which a program for realizing the method is recorded.

1 is an exemplary configuration diagram of a logistics information system to which the present invention is applied.

2 is a diagram illustrating an embodiment of a vehicle / cargo dispatch problem used in the present invention.

Figure 3 is an embodiment flow diagram for a vehicle dispatch method for logistics workflow according to the present invention.

4 is a conceptual illustration of a basic model used in the present invention.

5A to 5D are conceptual illustrations for making one path by connecting two points used in the present invention.

Figures 6a and 6b is an exemplary embodiment for the minimum cost problem used in the present invention.

* Explanation of symbols for the main parts of the drawings

11: Freighter 12: Transporter

13: Logistics Center

In order to achieve the above object, the present invention provides a vehicle dispatch method for a logistics workflow in a logistics information system, comprising: a first step of modeling a single garage, a single cargo type, a single vehicle type, and a plurality of demand sites as a basic model; A second step of generating a delivery route of a vehicle / cargo having a minimum cost based on the saving concept for the modeled basic model; And a third step of distributing an optimal vehicle according to the vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the minimum cost.

In addition, the present invention provides a vehicle dispatch method for a logistics workflow in a logistics information system, comprising: a first step of modeling a plural height, plural cargo types, plural vehicle types, and plural demand sites as an extension model; A second step of generating a delivery route of a vehicle / cargo having a minimum cost based on the saving concept for the modeled extension model; And a third step of distributing an optimal vehicle according to the vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the minimum cost.

The present invention for achieving the above object is a logistics information system having a processor, the function of modeling a single garage, a single cargo type, a single vehicle type, a plurality of demand sites as a basic model; Generating, for the modeled basic model, a delivery route of a vehicle / cargo having a minimum cost based on a saving concept; And a computer readable recording medium having recorded thereon a program for realizing a function of optimizing an optimal vehicle according to a vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the minimum cost.

In addition, the present invention provides a logistics information system having a processor, comprising: a function of modeling an plural garage, plural cargo types, plural vehicle types, plural demand sites as an extension model; Generating, for the modeled extension model, a delivery route of a vehicle / cargo having a minimum cost based on a saving concept; And a computer readable recording medium having recorded thereon a program for realizing a function of optimizing an optimal vehicle according to a vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the minimum cost.

The present invention is to implement a matching service that can save the logistics cost and time through the efficient connection between the transport demand of cargo generated in each region of the country and the transporter owning the vehicle as a means of transportation.

To this end, the present invention is to find the distribution path of the freight vehicle to minimize the input vehicle and minimize the tolerance rate when the information is stored in the database (DB) knowing the demand quantity, supply amount and location and capacity of the vehicle, etc. give.

The present invention provides a method for finding a delivery route of a plurality of cargoes supplied from a plurality of supply sites to a plurality of demand sites. First, the initial solution is obtained using the proposed matching method, and then the vehicle dispatch method using the saving matrix is performed.

By utilizing the present invention, it is possible to reduce the logistics cost of each company by reducing the transportation time and transportation costs incurred in the logistics industry and reducing the tolerance rate. It also helps rationalize the management of related companies by quickly and accurately generating dispatch plans.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is an exemplary configuration diagram of a logistics information system to which the present invention is applied.

An object of the present invention is to implement a matching service that can save the logistics cost and time through the efficient connection between the transport demand of the cargo generated in each region of the country and the transporter owning the vehicle as a means of transportation.

Matching service is a kind of distribution and transportation plan of vehicle and cargo. It is used to develop a distribution system of freight truck that minimizes transportation time, transportation cost and tolerance rate when the demand and supply of cargo and location and capacity of the vehicle are known. Can be implemented.

The relationship between the vehicle / cargo dispatch and delivery planning program proposed in the present invention and the entire logistics information system is shown in FIG. 1.

Referring to FIG. 1, the cargo owner 11 and the carrier 12 receive services through various user interface programs provided by the distribution center 13.

Here, the portion where the cargo owner 11 connects the cargo requested to be transported with the vehicle of the carrier 12 is a vehicle / cargo dispatch / delivery planning system.

The vehicle / cargo dispatch / delivery planning system selects a vehicle suitable for transporting each cargo and establishes an appropriate transportation plan of the vehicle to reduce the freight cost of the cargo and establish an efficient transportation plan of the carrier 12.

The situation of the vehicle / cargo dispatch problem to be presented in the present invention is shown in FIG.

As shown in FIG. 2, the problem situation is that a plurality of cargoes must first be delivered from a plurality of supply sites to a plurality of supply sites. In this case, each single cargo has one supply and one demand. Vehicles capable of delivering this cargo depart from a plurality of garages. The purpose of the delivery plan is to minimize the delivery price while minimizing the vehicle input.

The input information and output information of the vehicle / cargo dispatch and delivery method of the present invention are shown in Tables 1 and 2 below.

Classification Detailed Input Explanation Fixed data Distance Information Distance between each point or transit time (by departure time) All kinds of vehicles All kinds of cargo Cargo and Vehicle Suitability Types of Cars that Each Cargo Can Carry Working hours at each branch Allowable time for loading / unloading cargo into vehicle Change data Cargo Information Shipper, type, place of supply and demand of cargo, quantity of cargo, time of arrival of cargo, loading / unloading time, class of cargo Vehicle Information Borrower, location, type, capacity, upper limit of departure time, desired departure place, desired destination, desired price

Classification Detailed output Explanation Output related to cargo Origin and destination of cargo Departure time and arrival time of transport Departure and departure times of the cargo Hauling vehicle Vehicle-related output Cargo to carry Origin and departure time of origin Waypoint of vehicle Departure / Arrival Time of Each Stop Including Arrival News Cargo to get off at each stop Include cargo to drop off at destination Travel distance between each stop

Looking at the vehicle dispatch method for logistics workflow according to the present invention for solving the vehicle / freight dispatch problem presented in Figure 2 as shown in FIG.

As shown in FIG. 3, when data is input (301), vehicles are sorted by garage (302), and cargo is sorted by supply (303).

Looking at the vehicle route method by the heuristic method, first, the garage is arranged in the order close to the source of supply (304) and the cargo information is stored for each supply (305). Then, the saving value is calculated and the saving list is generated (306).

Thereafter, after sorting the saving matrix in descending order (307), the vehicle is allocated to the demand site in advance (308), and a single path is allocated to the remaining demand site (309).

Next, view the saving matrix, connect the paths (310), and generate new paths for all remaining points (311).

Finally, after storing the path (312) to output the solution, the solution is output (313).

Now, the minimum cost solution while satisfying the vehicle and cargo constraints described in Tables 1 and 2 will be described in more detail in two steps.

In this embodiment, we propose a solution from a model that simplifies the constraint and then expand it.

First, the basic model of the case of a single garage, a single cargo type, a single vehicle type, and a plurality of demand sites is considered as a basic model.

The problem situation of the basic model is shown in (Table 3).

Contents Remarks vehicle There is no restriction on the type of cargo for a single garage or multiple vehicles. Can think of a garage at the supply Cargo / demand Single Carrier, Single Supply Site Single Cargo Combination Possible types of vehicles No restrictions on cargo generation and arrival time The same room then delivered to multiple demands purpose Minimization of Input Vehicle Minimization of Total Travel Distance Reduce the number of vehicles using economy techniques

The core of the solution of the basic model is to create a vehicle route based on the concept of economy. Let's look at the savings concept.

As shown in FIG. 4, assuming that a vehicle is allocated at a starting point 1 to a vehicle at a point i and a point j respectively, and the vehicle is returned to the starting point, the moving distance of the vehicle at this time is “d _1i + d _i1 + d _ij + d _j1 ".

However, if points i and j are transported by one vehicle, the total travel distance at this time is "d _1i + d _ij + d _j1 ".

Total distance traveled with two vehicles "d _1i + d _i1 + d _ij + d _j1 " and total distance traveled with one vehicle at point i and point "d _1i + d _ij + d The difference of _j1 "is the saving value of point i and point j. This saving value means that when i and j are combined in the same path, the distance by "d _i1 + d _j1 -d _ij " is shortened and the number of moving vehicles is reduced. Therefore, in the basic solution, the values of the saving matrix between the points are sorted in ascending order, and if the constraints of vehicles, cargo, and points are satisfied from the pairs of points having the larger saving matrix, the two points are routed to one path. Include it in

Two points must belong to one of the following four cases in order to connect one point and make one path. Let R _k be the set of paths made so far.

First, when the two connected points are not included in the existing path, as shown in FIG. 5A, when two points to be connected are not included in any existing path, the two i and j are Be the first and last destination of the new route L.

Secondly, when only the point i is included in the existing route, as shown in FIG. 5B, when only the point i is included in the existing route L ₁ , only the point j and the point j are the final destinations in the route. Connection is possible. Then point j becomes the final delivery destination of route L ₁ .

Thirdly, when only the point j is included in the existing route, as shown in FIG. 5C, when only the point j is included in the existing route L ₁ , the point i and the point j are the first destinations in the route. Connection is possible. Then point i becomes the first delivery destination of route L ₁ .

Fourthly, if both points are included in the existing path, as shown in FIG. 5D, when the point i includes the existing path L ₁ and the point j is included in the existing path L ₂ , the point i is the path. Branch i and branch j are connectable only if L ₁ is the last destination and point j is the first destination in path L ₂ . At this time, two paths are reduced to one.

As described above, there is a connection possibility only when the points i and j correspond to the above four types. In such a case, if all the constraints of the vehicle, the cargo and the branch are satisfied, such as all constraints such as the arrival time of the vehicle and the cargo and the working hours of the branch, the connection between the two branches is established.

The solution of the basic model is as follows.

First, the saving matrix is calculated by calculating the savings between all two points according to (1).

Sij = di1 + d1j -dij, where dij is the distance from point i to point j,

i = 2, N

j = 2, N

In Equation 1, N-1 is the total number of demand sites, where point '1' means the supply site.

Second, list the savings listed in order of magnitude from which the savings are positive and dividing the two points that make those savings into a starting point and a destination point. Let NK be the total number of savings in the savings list.

Third, the demand Q _i of the delivery point is greater than the maximum loading capacity of the vehicles currently available, and the dispatchers more than one car to that point separately, and the remaining quantity to the requirement Q _i 'of that point.

Fourth, let K = 0.

Fifth, K = K + 1

If K> NK (i.e. read the savings list), then proceed to step 9.

The two largest savings points are read from the savings list. After the third course, go to the sixth course.

Sixth, if two points are bound together, all of the given constraints can be satisfied. In addition, it displays which path is included in each point.

Repeat steps seven and five to find all the paths.

Eighth, check which route all delivery points are in. If there is a point that is not included in the route, let that branch form a new route by itself. However, the point is subject to a third process.

If all delivery points are included in any route, proceed to step 9.

Ninth, a suitable vehicle is allocated according to the vehicle loading capacity according to the quantity of each route. It is finished when the vehicle has been assigned to all routes. If there are routes that cannot be assigned to vehicles, they are eliminated. In addition, in the eighth process, the route allocated solely to a single point is eliminated. Then, it is included in the missing paths and proceeds to the fourth step in order for the points to form a new path.

Now let's take a closer look at the extended model.

In the extended model, the multiple model, multiple feeder, desired starting point and desired destination constraints are added in the basic model.

In the baseline model, the garage and supply could be assumed to be at the same point because it is a single source and a single garage. However, when there are a plurality of garages and a supply source, when selecting a vehicle from each supply source, it is necessary to determine which vehicle in the garage to select. The first thing you can think of as a standard for choosing a garage is distance. In other words, the first choice is a garage close to the supply.

On the other hand, when the desired departure point of the vehicle is given, since the desired departure point can be regarded as the garage, it is not necessary to receive information about the garage in particular. The problem situation of the extended model is shown in FIG.

The solution to the extension model is as follows.

First, select feed source p = 1.

Second, arrange each garage in order of distance from point p.

Third, the vehicle is ordered according to the desired departure point with reference to the sorted garage.

Fourth, the basic model is executed. At this time, the criteria for selecting a vehicle uses the order of the vehicle determined in the third process.

Fifth, select the feed source p = p + 1. If there is no supply source to select, exit and proceed to the second step.

The solution of the extended model is shown in FIG. 3.

Meanwhile, the minimum cost initialization problem will be described in more detail.

Since it is difficult to obtain a vehicle dispatch that minimizes the cost in reality, the present invention considers the cost only when obtaining an initial solution of the vehicle dispatch. In other words, in the solution of the basic model, the basis for assigning the initial vehicle to each cargo was that the capacity of the vehicle was large.

The problem of matching vehicles and cargo that minimizes overall costs is transformed into a transport problem (a special case of the minimum cost problem). In other words, the supply points become vehicles and the demand points become cargo, where the origin and destination of the cargo are the same as the desired origin and desired destination of the vehicle and the transport conditions (time constraints, the combination of transportable vehicles and cargo) If there is a call between them, the cost of this call is "the price given by the vehicle plus the cost of the class of cargo".

Thus, after converting to a transportation problem, a solution is obtained using a solution of a general minimum cost problem (network simple method).

However, this solution cannot be used as it is. Because if the vehicle has multiple desired origins and desired destinations, the year in which a vehicle carries several cargoes may come out, which may be impossible in practice. Therefore, the solution must be analyzed to eliminate these matches. There is then a cargo to which the vehicle is not assigned, which assigns the vehicle by empirical means. That is, for vehicles that are connected to the calls coming in as cargo, the cars with a high flow amount flowing in the call are preferentially assigned.

In the case of the presence of vehicles and cargoes, there are five cases of switching to the minimum cost problem (ie, when the supply is more than the demand, when the demand is more than the supply, the supply and demand are balanced, but the demand nodes are individually If it is not met, the supply and demand are balanced, but if the supply node cannot send the supply individually, and the supply and demand are both balanced, the supply and demand for each supply node and demand node can be met. Case) occurs.

Other than the fifth of the five problems above, other problems become impossible. Thus, there is a need for a method of converting these impossible problems into possible problems. To deal with all five of the above and make it a problem, you need to add processed feed and hydrolyzed ingredients.

For example, suppose you have vehicles and freight, as shown in Tables 4 and 5.

Information about the vehicle Number of vehicle Capacity of vehicle Hope origin Hope destination One 10 One 2 2 20 2 3 3 30 3 4 4 40 2 3

Information about the cargo Number of cargo The volume of cargo Hope origin Hope destination One 10 One 2 2 50 2 3 3 30 2 3 4 10 3 4

Creating a network of minimum cost problems with information about vehicles and cargoes in Tables 4 and 5 results in a network as shown in FIG. 6A.

Referring to FIG. 6A, since the total amount of supply nodes is 80 and the total amount of demand nodes is 100, there is an unbalanced problem.

Even if a processing paper with a capacity of 20 is created to create a balance problem, it is a question of which demand node to connect. In addition, even for each demand node, a node having a demand amount of 50 cannot be satisfied since the sum of the supply amounts of supply nodes connected thereto is 40. Therefore, in such a case, it is necessary to make the processing paper having the total demand of the demand node as the supply capacity.

Consider the case of a supply node with a supply amount of 30. The node associated with this node has a demand of 10, and after allocating 10 to this node, the remaining supply 20 remains. In order to deal with this residual 20, a hydrogel must be made. The demand for hydrolyzate can be obtained by calculating the remaining supply of all supply nodes and summing them up, but the overall problem must be balanced, so the supply of all supply nodes is the sum of all the supply nodes. In this case, since the processing node also calculates the sum of all the demands, the overall problem becomes a balance problem, and all the supply nodes can send the remaining supply to the water supply point, so any problem becomes a possible problem. That is, the network is changed to the network shown in FIG. 6B.

After this matching, one supply node and several demand nodes are connected. Considering the supply node as the vehicle and the demand node as the cargo, only results that can be matched according to the desired origin and destination of the cargo and the desired origin and desired destination of the vehicle are produced. However, since the desired starting point and the desired destination are different for each cargo, even if it is concluded that the vehicle can be matched with multiple cargoes, it may not be possible in practice. Therefore, in this case, among the cargoes assigned to the vehicle, the cargo with the highest volume is assigned first, and the cargoes of which the desired origin and the desired destination of the cargo coincide can be allocated.

Assignment in this way results in unassigned vehicles, unallocated cargo, and cargo that has not met all the demand. In this case, unassigned vehicles, cargo with remaining demand, and unassigned cargo are allocated one by one, starting with a high flow volume considering the travel time between desired origin and desired destination and cargo's desired destination. Take

Now, let's look at how to make possible arcs.

To solve the minimum cost problem, information about the start, end, and cost of the call, the lower limit of the call, and the upper limit of the call are entered. Here, if an arc is made for all supply nodes and demand nodes, the number of arcs is enormously large. That is, even if the desired starting point and the desired destination of the vehicle and the cargo do not coincide with each other, a call is generated, and thus an inefficiency in the memory appears. Therefore, a call is made around a vehicle-cargo pair in which the desired starting point and the desired destination of the vehicle and the desired starting point and the desired destination of the cargo coincide.

First, select a pair in which the desired starting point of the vehicle and the desired starting point of the cargo are the same.

Second, compare the "available time of the vehicle + travel time between the desired starting point and the desired destination" and "required arrival time of the cargo to the desired destination".

If " available time of the vehicle + travel time between the desired starting point and the desired destination " " requested arrival time of the cargo to the desired destination "

On the other hand, in the case of " available time of vehicle + travel time between desired starting point and desired destination " " required arrival time of cargo to desired destination "

Third, it is determined whether the arrival time required for the desired destination of the cargo is later than the start time of the destination. If it is late, proceed to the fourth step. On the other hand, since the flight arrives earlier than the working start time of the destination, no call is made.

Fourth, if the available time of the vehicle is earlier than the occurrence time of the cargo, the available time of the vehicle is regarded as the occurrence time of the cargo and before the working deadline of the desired destination of the cargo in consideration of the time from this time to the arrival point of the cargo. See if the cargo can arrive. If it is possible to arrive earlier, the cost of the call will be calculated at the price presented to the vehicle, or no call will be made.

By converting the information about the vehicle and the cargo to the minimum cost problem as described above and solving the problem, the matching result of the vehicle and the cargo can be obtained in consideration of the cost. This matching result is the initial solution of the basic model.

The matching method combined the initialization method to solve the minimum cost problem is as follows.

First, select the cargo type q = 1.

Second, select vehicles that can carry cargo type q.

Third, the vehicle dispatch and cargo delivery problem for the selected cargo and vehicle is transformed into a transportation problem (minimum cost problem). At this time, the transportation problem is configured as follows.

The supply point is the vehicle and the demand point is the cargo. And if the desired starting point and destination of the vehicle coincide with the starting point and destination of the cargo, there will be a call, and the cost will be "the price presented by the vehicle plus the cost of the class of cargo".

Fourth, solve transportation problems.

Fifth, eliminate non-matching possible matches.

Sixth, for each cargo that is not assigned a vehicle, it matches the vehicle with the least cost among the vehicles that can transport this cargo.

Seventh, select cargo type q = q + 1. Exit if there is no cargo type to choose.

The solution obtained by the above method becomes the initial solution of the basic model.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the spirit of the present invention for those skilled in the art to which the present invention pertains, and the above-described embodiments and accompanying It is not limited to the drawing.

The present invention as described above, there is an effect that can reduce the logistics cost of each company by reducing the transportation time and transportation cost and the tolerance rate in the logistics service. In addition, the efficient combination of cargo and vehicles increases the profits of shippers and owners, and rationalizes the management of related companies by automatically and quickly establishing dispatch plans. On the other hand, it is possible to create a mathematical model of the problem of vehicle dispatch in real life and to help develop a solution. In addition, the technology for solving the vehicle dispatch problem can be applied to problems occurring in a similar industry, for example, aircraft scheduling, ship scheduling, etc., and more complex and new vehicle dispatch model using the vehicle dispatch model proposed by the present invention. There is an effect available for developing solutions.

Claims (5)

In the vehicle dispatch method for logistics workflow in the logistics information system, A first step of modeling a single garage, a single cargo type, a single vehicle type, and multiple demand sites as a basic model; A second step of generating a delivery route of a vehicle / cargo having a minimum cost based on the saving concept for the modeled basic model; And A third step of dispatching an optimal vehicle according to the vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the least cost Vehicle dispatch method for logistics workflow comprising a. In the vehicle dispatch method for logistics workflow in the logistics information system, A first step of modeling a plurality of garages, a plurality of cargo types, a plurality of vehicle types, and a plurality of demand sites by an extension model; A second step of generating a delivery route of a vehicle / cargo having a minimum cost based on the saving concept for the modeled extension model; And A third step of dispatching an optimal vehicle according to the vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the least cost Vehicle dispatch method for logistics workflow comprising a. The method of claim 2, The expansion model, Substantially, for a logistics workflow characterized by the addition of multiple garages, multiple feeders, desired starting point and desired destination constraints to a basic model with constraints of single garage, single cargo type, single vehicle type, and multiple demands. Vehicle dispatch method. In the logistics information system with a processor, Modeling a single garage, a single cargo type, a single vehicle type, and multiple demand sites as a basic model; Generating, for the modeled basic model, a delivery route of a vehicle / cargo having a minimum cost based on a saving concept; And A function of dispatching the optimal vehicle according to the vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the minimum cost A computer-readable recording medium having recorded thereon a program for realizing this. In the logistics information system with a processor, A function of modeling plural heights, plural cargo types, plural vehicle types, plural demand sites as an extension model; Generating, for the modeled extension model, a delivery route of a vehicle / cargo having a minimum cost based on a saving concept; And A function of dispatching the optimal vehicle according to the vehicle loading capacity according to the quantity of each delivery route of the vehicle / cargo having the minimum cost A computer-readable recording medium having recorded thereon a program for realizing this.