JP2988719B2 - Assignment device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention relates to an element (for example, a job) belonging to a first set.
Is assigned to elements (for example, resources) belonging to a second set.

(Prior Art) In recent years, various types of allocating apparatuses have been developed which perform computer resource processing for allocating resources to jabs, such as allocating crews for various transportation operations and allocating work, and the like. This type of assigning device basically assigns a plurality of jobs sequentially according to certain criteria.
The job is sequentially assigned to an appropriate resource while taking out the jobs one by one according to the order. At this time, the job ordering may be reviewed during the assignment process,
In addition, various attempts have been made to develop a function of changing the assignment state in order to find a better assignment if an appropriate resource that can be assigned to a job is not found.

However, in a conventional apparatus that solves the assignment problem by ordering jobs, if the number of jobs and resources becomes enormous, the assignment process takes a lot of time or becomes impossible as the assignment process proceeds. It becomes easy to fall into a state. Moreover, a problem that a better assignment result is hardly obtained is likely to occur.

In other words, since the job is ordered and the assignment process is performed, a relatively appropriate resource can be assigned to the first job to be processed, but it is difficult to find a suitable resource to assign to a later job. Becomes For this reason, there have been various problems in performing the allocation process efficiently, such as the need to frequently change the allocation state.

(Problems to be Solved by the Invention) In such a conventional allocating apparatus, a plurality of jobs are ordered in advance according to a predetermined rule, and resources are allocated according to the order. When the number of data items increases, there is a problem that the processing becomes very difficult, and it becomes difficult to obtain a better allocation result.

The present invention has been made in view of such circumstances,
An object of the present invention is to provide an assignment apparatus that can execute assignment of a plurality of jobs and a plurality of resources at high speed and efficiently, and can easily obtain a better assignment result.

[Configuration of the Invention] (Means for Solving the Problems) The present invention provides at least main body information of a plurality of elements belonging to a first set and information indicating a starting state of the elements,
A first set element database in which information indicating an end state is stored, a second set element database in which information of a plurality of elements belonging to a second set is stored, and an element belonging to the first set. A constraint database storing constraints relating to assignment of information belonging to the second set to elements, and belonging to the first set stored in the first set component database according to constraints of the constraint database Assignment determining means for selectively assigning elements and information on elements belonging to the second set stored in the second set element database, wherein the assignment determining means belongs to the first set The noted element and the other element according to the information indicating the end state of the noted element of the plurality of elements and the information indicating the beginning of the other element Combined into one, the set of the one for wearing meta element is characterized in that assigned to one of the elements belonging to the second set in accordance with the above constraints.

(Operation) According to the present invention, a plurality of elements belonging to the first set are put together as a set of elements in accordance with information respectively representing the start and end states of each element, and the set of put elements is used as a unit. Since it is assigned to one element belonging to the second set,
For example, it is possible to assign a plurality of jobs to the same resource prior to making a detailed assignment between the job and the resource. That is, prior to individually assigning jobs and resources, a set of jobs having a predetermined status relationship is assigned to resources. In addition, it is possible to prevent the plurality of allocation processes from being performed, and to efficiently perform the allocation process in stages.

As a result, apparently, the number of jobs is reduced to prevent the allocation process from becoming complicated, and it becomes possible to perform detailed allocation after allocating a set of jobs to one resource. It is possible to reduce the possibility of occurrence of a new allocation situation, shorten the time required for the allocation process, and easily and efficiently obtain a better allocation result.

(Embodiment) Hereinafter, an assignment device according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows a schematic configuration example of the apparatus of the embodiment.
Reference numeral 1 denotes a job database that stores information about a plurality of jobs that are elements belonging to a first set, and reference numeral 2 denotes a resource database that stores information about a plurality of resources that are elements that belong to a second set. The information on a plurality of jobs stored in the job database 1 includes start information indicating the start state of the job and end information indicating the end state of the job, in addition to body information indicating the content of the job. I have.

Reference numeral 3 denotes a constraint database storing various constraint conditions regarding the assignment of the job and the resource. The judgment processing unit 4 refers to various constraint conditions stored in the constraint database 3 and assigns jobs and resources stored in the job database 1 and the resource database 2 to a plurality of jobs. It has a function of allocating a set of jobs to one resource as a unit.

The allocation determining unit 5 forming the main body of the apparatus executes a detailed allocation process between each job and the resource while referring to the allocation result of the job set and the resource allocated by the determination processing unit 4 described above. The assignment determining unit 5 determines the assignment state of each job and each resource.

As shown in FIG. 1, the apparatus of this embodiment includes a judgment processing unit 4, and prior to the assignment deciding unit 5 deciding in detail the assignment between the job and the resource, the judgment processing unit 4 is used.
Generates a job set in which a plurality of jobs are put together, and allocates the job set (a plurality of jobs) to resources, thereby reducing the apparent number of jobs and efficiently performing the allocation processing. It is characterized by having such a configuration.

The characteristic processing functions of the determination processing unit 4 will be described below. For example, as shown in FIG.
When considering a transportation business connecting a point, the operation is planned on the basis of time, for example, as represented in a diagram as shown in FIG. Specifically, in the example shown in FIG. 3, point A departs at [9:00] and moves to B at [9:30].
The operation a reaching the point and the operation b departing from the point B at [10:00] and arriving at the point A at [10:30] are respectively planned as jobs in the transportation business.

Information on such a job is stored in the job database 1 in a format as shown in FIG. 4, for example. In the example shown in FIG. 4, the information about the job includes the departure point, departure time, arrival point, arrival time,
Each type of information (type of vehicle used for transportation business) is specified and expressed.

In addition to such information, in the constraint database 3, information on the above-mentioned two locations is attached to each job as information attached to each job. As shown in FIG. 5A, the location A is "garage". , B point is described as "not a garage". The information on the status of “garage” indicates, for example, that there is an advanced maintenance device for the machine, that there is an excellent rest facility for personnel, and the like. Information on such maintenance equipment and rest facilities may be expressed separately from the information such as "garage" described above,
Needless to say, the types of the facilities may be further finely classified, and the information on the presence or absence of these may be individually described.

The “minimum stay time”, which is the additional information shown in FIG. 5 (b), is “the minimum time to stop at the point when a certain resource is allocated to a certain operation and arrives at the point”.
Is represented. The auxiliary information "minimum stay time" is information for securing the maintenance time of the machine, the break time of the personnel, and a margin time for a delay when the operation does not proceed as planned. Of course, it is needless to say that these times may be defined separately.

Now, as resources for the above-described job, there are machines and personnel in this example. Now, it is assumed that a machine is considered as a resource, and which machine is to be assigned to the operations a and b, which are the jobs described above. By the way, in the past, the machine was assigned for each operation a, b, but in this embodiment, the two resources are combined into one under the above-described determination processing unit 4, and the same resource (machine) is assigned. An assignment is determined.

FIG. 6 shows this processing procedure. First, a plurality of jobs are grouped into one unit based on information indicating the start state and the end state of the operation (job). Will be That is, the first job (operation)
In this example, a point "not in the garage" is searched (step a1). In this case, point B is determined. Thereafter, the following processing is repeatedly executed for all search results of the points determined as not being a garage (step a2), and when the processing is completed, the above-described job set generation processing is completed (step a2). Step a3).

The process for a point determined to be not in the garage is started by first arranging the operations that arrive and depart from the point B (step a4). The organization of this operation (job)
The information of each job stored in the above-mentioned job database 1 is checked, and a list of operations that arrive and depart point B is created as [9:30, arrival, operation a] [10:00, departure, operation b], etc. This is done by doing Thereafter, referring to this list, the following processing for all departures departing from point B is repeated, and when the processing is completed, the same processing is executed for other points (step a5). .

The processing for all the departures departing from the point B is that the arrival (time) between the arrival time of the arrival operation and the arrival time of the operation arrives at the point B before a certain operation is described above. Find out what is longer than the minimum stay time (step a6). If there is no arrival service that satisfies such conditions, the process proceeds to the next operation without doing anything. If only one service that satisfies the above conditions is found,
The same resource (machine) as the departure service that is the processing symmetry
(Step a7). In this example,
Since the arrival service a satisfies the above-described conditions for the departure service b, it is determined that these services a and b can be allocated to the same resource (machine).

In addition, there are a plurality of arrival services (two or more) that satisfy the above conditions.
If found, several decisions are possible according to the number. For example, a separate decision is made according to the following criterion, and a departure service that can be allocated to the same resource as the departure service is determined (step a8). . The algorithm for this determination is as follows: [1] For each departure operation to be processed, which arrival operation is to be allocated to the same resource,
No judgment is made at this time.

[2] It is determined that the so-called FIFO flow is to be departed early from the service that arrived earlier. Specifically, as shown in FIG. 7, an earlier arrival service is assigned to an earlier service. According to such a determination algorithm, the stay time at the point B can be substantially equalized. The seventh
The figure shows that the operation connected by the lone line at the point B is determined as a plurality of jobs to be allocated to the same resource.

[3] Conversely, it is also possible to determine that a service that arrives early departs later in the FILO style. According to such a determination algorithm, a combination of operations as shown in FIG. 8 is obtained, and the stay time at the point B changes variously. As a result, it is possible to secure a long stay time for a small number of resources, but to spend that time on maintenance of machinery,
It can be used as a break time. By the way, if the time required for such maintenance and personnel meals and breaks is one hour, it is difficult to secure that time with the FIFO flow judgment algorithm described above.
According to the FILO flow determination algorithm, the above time can be relatively easily secured.

In the case where the number of departures is small as shown in FIG. 9, there is a large difference between the above-described FIFO flow determination algorithm and the FILO flow determination algorithm. Such an operation (job) schedule often occurs in the evening in the usual transportation operation, where resources (machines and personnel) stay at the point B and the resources depart the next morning. The operation (job) schedule for the next day is as shown in, for example, FIG. 10, so that the operation schedule throughout the day can be regarded as equal to the number of sleds.

According to the above-described FILO flow determination algorithm for the operation schedule shown in FIG. 9, the constraint time of the two resources is as follows: resource 1… 9:00 to 9:30 (30 minutes) resource 2… 9: From 15 to 10:45 (1 hour 30 minutes), a total of 2 hours of restraint time occurs. On the other hand, according to the above-described FIFO flow determination algorithm, the constraint time of the two resources is as follows: resource 1 9: 0 to 10:45 (1 hour 45 minutes) resource 2 9:15 to 9:45 ( 30 minutes), giving a total of 2 hours and 15 minutes of restraint time. So 15
There is a difference in the minute constraint time. And in general, there is usually a constraint on the daily restraint time for personnel, and it is desirable to shorten this as much as possible.
A large difference occurs depending on which judgment algorithm is used.

It should be noted that the operation schedule at the time of departure shown in FIG. 10 also needs to be determined in the same manner as to when to depart the staying resource.

[4] It is also possible to appropriately combine the above-described FIFO and FILO determination algorithms. While the FIFO flow determination algorithm generally has the advantage of being able to equalize the resource constrained time, the FILO flow determination algorithm has significant advantages in certain situations, for example, with the above-mentioned staying. Therefore, the algorithm for determining the FIFO flow is not always employed, and the above-described FI is used only in specific situations.
An algorithm for determining the LO flow may be employed.

By appropriately adopting the above-described determination algorithm, a plurality of jobs are combined into one, and a set of the jobs is assigned to the same resource, so that the apparent number of jobs is reduced and the assignment processing is performed. It is possible to proceed efficiently.

In the above description, the minimum stay time has been described as being constant regardless of the time, but the minimum stay time may be varied according to the time zone. For example, the minimum stay time may be varied according to the departure time of the service.

Also, it is of course possible to make a determination that two jobs are allocated to the same resource at all points.
In this case, as a result of the overall determination, it is determined that more jobs, for example, a job from point A to point B and a job from point B to point A are allocated to the same resource. become.

Further, when there are different types of jobs, such as a large bus and a small bus, for example, the allocation to resources is determined by focusing only on jobs of the same type. Good. At this time, it is of course possible to change the minimum stay time according to the type of job. In this case, the information about the job shown in FIG. 5 described above may be given in more detail according to the type.

Further, in the above-described embodiment, it is assumed that all the departures and arrivals at point B are performed between point A and point B.
In a case where the operation of departure and arrival at a point is an operation from a plurality of points, it is possible to make a determination on allocation to resources in the same manner.

FIG. 11 shows an example of a transportation business connecting points A and B and between points A and C. In the transportation business shown in FIG. 11, it is assumed that there is only an operation from point A to point B and an operation from point C to point A as shown in FIG. In such a case, for example, information as shown in FIG. 13 is stored in the job database 1 as the additional information on the points, and the processing is executed while identifying the points B and C, and Note that it takes 25 minutes to move between points B and C.

If such attribute information is given, for example, the stay time at the points B and C is captured as the minimum stay time plus the time required for the movement between the points B and C (resource forwarding time). Processing can be performed in the same manner as in the embodiment described above. In other words, the minimum stay time on the operation is as follows: (a) arrive at point B, depart from point B (b) arrive at point C, depart from point C (c) arrive at point B, depart from point C (d) Consider arriving at point C and departing from point B into four cases, create a set of jobs to satisfy the minimum staying time conditions for each of these cases, and assign them to the same resource. Good.

In this case, it is needless to say that the assignment between the job and the resource may be determined by separately considering the time required for the movement between the points B and C (the forwarding time of the resource).

By the way, in the above-described embodiment, depending on the situation, it may be determined that an operation in which departures and arrivals at the point B are separated for a long time is allocated to the same resource. That is, since only the minimum stay time is specified, it may be determined that an operation (job) that is long apart in time is allocated to the same resource.

However, when there is no place to keep resources for a long time at the point B, or when the allocation status of other resources at that time is considered, a problem occurs that the above-described determination of allocation is not appropriate.

In such a case, if the information on each point is given as shown in FIG. 14, for example, and the maximum stay time is also specified, the above-mentioned problem can be solved. That is, in this case, the vehicle arrives at the point B before a certain departure, and the time (time) between the arrival time of the departure operation and the departure time of the departure is equal to or longer than the minimum stay time described above. , And those whose staying time is within the maximum staying time, so that these operations may be obtained as one set.

As described above, in the case where the stay time between two operations obtained as one set of jobs forming a group exceeds the minimum stay time, the time exceeding the minimum stay time is a certain amount. In a sense, it can be said to be wasted time. In order to eliminate such wasted time, for example, it is only necessary to take measures such as (a) delaying the operation arriving at the point B, (b) accelerating the operation departing from the point B, and (c) combining these. In order to perform the process (c), it is easiest to delay the operation arriving at the point B and at the same time advance the operation departing from the point B.

In addition, for example, (d) when there is a restriction on the operation time, the restriction is obeyed. (E) When there is something desirable at the operation time, for example, the earliest operation and the latest operation are followed. The departure and arrival times are not changed for the operation, and the departure and arrival times are changed only for other operations.

Specifically, when there is a restriction on the operation time, the following may be performed. For example, when the restrictions on the operation time in the operation schedule shown in FIG. 3 are given as shown in FIG. 15, for the two operations determined to be allocated to the same resource in the previous embodiment,
What is necessary is just to perform the determination processing as shown in the figure.

That is, 2 which is determined to be allocated to the same resource
First, an extra stay time at the point B is calculated for one operation (step b1). Then, if the extra stay time is not [0] (step b2), it is determined how much the arrival service can be delayed and how fast the departure can be made (step b2).
b3, b4), the time during which the arrival service can be delayed is determined as x, and the time during which the departure service can be advanced is determined as y. The calculation of the times x and y is performed with reference to the information shown in FIG. 15 described above.

Thereafter, it is confirmed that both the times x and y are not [0] (step b5), and the time to delay the arrival is set to [x / (x
+ Y)], and the time to accelerate the departure is [y / (x +
y)] (steps b6 and b7). still,
If the times x and y are both [0], the time is not changed.

By the way, it is also possible to reduce the waste of the above-mentioned stay time by earliering arrival service or delaying departure service. For example, when the constraint time for the operation schedule shown in FIG. 3 is given as shown in FIG. 17, when the useless stay time for the operation group is calculated according to the processing procedure shown in FIG. 10], [y = 1
0], and it is required that the arrival service be delayed by 5 minutes and the departure service be advanced by 5 minutes at the same time. However, such a change in the departure / arrival time does not satisfy the constraint time shown in FIG. 17, and therefore cannot be adopted.

Therefore, in such a case, first, a range in which the departure time for the operation a can be changed is checked. Then, as shown in FIG. 18, for operation a, [0: 00-9: 00] and [9:10]
~ 24: 00], it is required that the departure time can be changed. In the same manner, the range in which the departure time can be changed for the operation b is checked, and the sum of the required time (30 minutes) for the operation a and the minimum stay time (20 minutes) at the point B (50 minutes) is determined from the time range. ) Is subtracted, and this is determined as the departure time width that can be changed for the operation b as shown in FIG. That is, the changeable time width of the departure time for the operation b is obtained in association with the operation a.

Then, the time that satisfies the constraints of the operations a and b is represented by two lines as the departure time of the operation a as shown in FIG. 18, and the departure time of the operation a is satisfied at the time that both of these are satisfied. It is required that the time be changed. In this example, the third
In the operation schedule shown in the figure, the departure time of the operation a is [9:00], and it is desirable to reduce the waste of the stay time without changing it as much as possible. Therefore, in this example, the operation a is not changed. If the departure time of the operation b is changed to [9:50], the above-described object can be achieved.

By the way, as an extreme example of eliminating the waste of the above-mentioned stay time, it is also possible to change the operation time so that the stay time at the point B becomes the minimum stay time. This is advantageous when the time of operation is not naturally given.

That is, if attention is paid to the operation between the two points A and B, the point A
In many cases, the number of operations from the point to the point B and the number of operations from the point B to the point A are equal. Accordingly, in such a case, for example, according to the information on the operation from the point A to the point B, the operation from the point B to the point A is determined so that the stay time at the point B becomes the minimum stay time as described above. What is necessary is just to generate automatically.

Regarding the resource staying at the point B, the operation from the point B to the point A may be entertained in the morning of the next day in accordance with the constraint condition of the operation time. In this case, the operation time may be changed by the intervention of the operator.

As described above with respect to some embodiments, according to the present invention, two jobs are combined into one based on the end situation and the start situation of a plurality of jobs, and this set of jobs is Is assigned to At this time, the information of the jobs to be combined into one is appropriately changed in accordance with the constraints on the jobs, and a job set without waste is generated. Therefore, even when the number of jobs is enormous, it is possible to reduce the apparent number of jobs and efficiently allocate resources. As a result, compared to the case where multiple jobs are ordered and assigned to individual resources,
In addition, there can be provided a great effect in practical use, for example, it is possible to efficiently proceed with the assignment process while systematizing the assignment status.

Note that the present invention is not limited to the above-described embodiment. For example, three or more jobs can be combined into one and assigned to the same resource. Also, various modifications can be made as to which point is used as a basis for grouping jobs. In addition, the present invention can be variously modified and implemented without departing from the gist thereof.

[Effects of the Invention] As described above, according to the present invention, the process of assigning a plurality of jobs to the same resource is executed before the detailed assignment of each job and resource is performed.
The number of jobs that can be apparently reduced reduces the time required for the allocation process, and reduces the possibility of an undesired allocation situation, thereby enabling a large amount of practical allocation. The effect is achieved.

[Brief description of the drawings]

FIG. 1 shows an assignment device according to one embodiment of the present invention, and FIG.
FIG. 2 is a diagram showing an example of a transportation business used in the embodiment, FIG. 3 is a diagram showing an example of a job in the transportation business, FIG. 4 is a diagram showing an example of the contents of a job database, and FIG. FIG. 6 is a diagram showing an example of a processing procedure in the apparatus of the embodiment, and FIGS. 7 to 10 are diagrams showing examples of combinations of jobs assigned to the same resource. FIG. 11 is a diagram showing another example of the transportation business, and FIG.
FIG. 13 is a diagram showing an example of the contents of a job in another transportation business shown in FIG. 13, FIG. 13 is a diagram showing another example of contents of a constraint database,
14 and 15 are diagrams showing still another example of the contents of the constraint database, FIG. 16 is a diagram showing a flow of a processing procedure for changing the time of the job, and FIG. 17 is a diagram of another constraint database. FIG. 18 is a diagram showing an example of contents, and FIG. 18 is a diagram showing an example of detection of a changeable time for changing the time of a job. 1 ... job database, 2 ... resource database, 3 ... constraint database, 4 ... judgment processing unit, 5 ...
... Assignment determination unit.

Claims (2)

(57) [Claims] 1. At least body information of a plurality of elements belonging to a first set and information indicating a starting state of the elements,
A first set element database in which information indicating the end state is stored; and a second set element information in which information of a plurality of elements belonging to the second set is stored.
A set element database, a constraint database storing constraints relating to assignment of elements belonging to the first set and information of elements belonging to the second set, and the first And assignment determining means for selectively allocating elements belonging to the first set stored in the set element database and information of elements belonging to the second set stored in the second set element database. The assignment determining unit may determine the target element according to the information indicating the end state of the target element of the plurality of elements belonging to the first set and the information indicating the start of another element. The other elements are combined into one, and a set of the combined elements is assigned to one element belonging to the second set in accordance with the above constraint. Allocation device that. 2. The method according to claim 1, wherein the assigning unit determines a start state of the noted element as the other element combined with the noted element of the plurality of elements belonging to the first set. 2. An element whose end state is the start state of the noted element is searched from the first set element database based on the information shown.
Assignment device as described.