JPS63163671A - Deciding and supporting system for assignment - Google Patents

Abstract

PURPOSE:To shorten an operation time by preparing plural algorithms for the procedure of an assignment decision, selecting the algorithm which decides the suitable assignment decision according to the conditions of the assignment decision, and reducing the frequency of standstill in the course of the assignment decision of the combination of plural jobs and plural resources. CONSTITUTION:When the assignment of the jobs and the resources is decided, the plural algorithms which decide the assignment are previously prepared, and among them, the algorithm which advances the assignment decision for the jobs and resources is selected by referring to the conditions in the course of the assignment decision and information on the jobs and the resources. Therefore, even when the conditions of the assignment decision is changed in the course of the procedure of the assignment decision, the procedure can be continued by using the algorithm which complies with the conditions. Thus the frequency of the standstill in the course of the assignment decision and the alteration of the assignment decision can be reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to an allocation decision support method that can effectively allocate combinations between a plurality of jobs and a plurality of resources.

(Prior Art) Conventionally, it has been common practice to allocate combinations of a plurality of jobs and a plurality of resources under various constraint conditions. For example, the allocation of workers (resources) to various tasks (jobs) is determined according to working conditions and working abilities (constraints).

Conventionally, such allocation decisions have been made solely by humans, but as the jobs and resources become more complex, manual allocation decisions become more difficult. In addition, in the process of proceeding with the allocation decision, conflicts tend to occur, such as, for example, resources that can be allocated to a certain job have already been allocated to another job, resulting in no resources to be allocated to that job, resulting in a deadlock in the allocation decision process. There are many cases. In this case, it becomes necessary to repeat the allocation determination procedure. For this reason, it cannot be denied that a large amount of time is required to determine the allocation of jobs and resources.

Therefore, recently, it has been considered to process this type of allocation decision using a computer.

FIG. 7 shows an example of a general processing procedure for determining separate allocation using a computer according to the prior art.

This process includes, for example, sequentially extracting a plurality of jobs from the first registered job (step bi), and sequentially extracting resources that can be allocated to the extracted jobs from among the plurality of resources (step b2). )
, the allocation constraints, etc. are checked and determined (steps b3 and b4). Specifically, this check is performed by checking the allocation constraint conditions (step b3) and then further checking the dynamic conditions (
Step b4).

Then, it is decided to allocate, for example, the first resource found among the resources that satisfy the allocation constraints etc. to the job (step b5), and the allocation status is stored (step b6).

Note that if a resource that can be allocated to a job is not found, the allocation is changed as appropriate.

Such a processing procedure is repeatedly executed for all jobs while removing jobs and resources for which allocation has already been determined, and the allocation is completed.

However, in such a processing procedure, resource allocation decisions are simply made sequentially starting from the first job, so it is easy to control the allocation decision process, but on the other hand, it is easy to get stuck in the process of the allocation decision procedure, and it is often difficult to do so. The problem is that an allocation change is required.

Therefore, recently, for example, as shown in Figure 8, which shows the flow of the processing procedure, in descending order of allocation constraints,
The idea is to proceed with the allocation decision.

That is, this processing procedure first investigates the ill convention for assignment of jobs and resources (step c).
i). Step C2), which sequentially extracts jobs with the strictest allocation constraints according to the results of this investigation;
Resources that can be allocated to the retrieved job are determined by checking the allocation constraints and the like for the resources that are sequentially retrieved from among the plurality of resources (step C3) (steps c4 and c5). This check is performed by checking the constraint conditions of the allocation (step C4) in the same way as in the previous allocation determination procedure, and then further checking the dynamic conditions (step c5).

Then, from among the resources that satisfy the allocation constraints, etc., the resource whose constraints are considered to be the most severe is allocated to the job (step C6), the allocation status is memorized (step C7), and all of the above processes are performed. This is done by repeatedly executing the job.

This processing procedure has the advantage that, compared to the processing procedure exemplified above, the allocation determination procedure is less likely to stall, and the allocation determination is changed less frequently.

However, in such a processing procedure, it is necessary to investigate the severity of the constraint conditions in advance, and
It is necessary to control job withdrawal while determining the severity of the treaty. Therefore, it is undeniable that not only the processing load on the computer increases, but also the processing speed decreases.

(Problem to be Solved by the Invention) In this way, in the past, when determining the allocation between multiple jobs and multiple resources, the result depends on the policy based on the allocation determination procedure. tdJ for allocation decision procedure! There are problems in that the if is complicated, many procedures are required to change the allocation determination, and furthermore, the allocation determination procedure requires a large amount of processing time.

The present invention has been made in consideration of these circumstances, and its purpose is to reduce the frequency of deadlocks in determining the allocation of combinations of multiple jobs and multiple resources, and to reduce the processing burden. Another object of the present invention is to provide a v1 guess determination support system that can shorten the processing time.

[Configuration of the Invention (Means for Solving Problems)] The present invention prepares a plurality of allocation determination algorithms when determining the allocation between jobs and resources according to the allocation constraint conditions for the jobs and resources. Then, in the process of determining the assignment of υ1, select which algorithm to proceed with determining the assignment of the job and the resource from among the aforementioned assignment determination algorithms by referring to the assignment status and the information of the job and the resource. It is characterized by:

In other words, from among a plurality of allocation decision algorithms, an allocation decision algorithm considered to be the most appropriate is selected based on the allocation decision situation during the allocation decision process, and the allocation decision procedure is proceeded with.

(Operation) According to the present invention, in the process of determining the allocation between a plurality of jobs and resources, an appropriate allocation decision algorithm is selected according to the allocation situation, and the allocation decision procedure is performed according to the selected allocation decision algorithm. will proceed.

Therefore, even if the status of the allocation decision changes during the process of proceeding with the allocation decision procedure, it is possible to proceed with the allocation decision procedure by adopting an algorithm that suits the situation. It is possible to reduce the occurrence of clogging and to reduce changes in allocation decisions.

Furthermore, when proceeding with the allocation decision procedure depending on the severity of constraints, for example, it is only necessary to judge the severity of the constraints according to the situation at that time, so the process does not become unnecessarily complicated. . Therefore, the processing load for determining allocation can be significantly reduced, and the processing can be performed efficiently and in a short time.

(Embodiment) Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a schematic diagram of an information processing system constructed by applying an embodiment of the present invention. A plurality of jobs to be subjected to allocation processing and a plurality of resource information for the jobs are stored in a job database 1 and a resource database 2, respectively.

The plurality of jobs registered in this job database 1 are, for example, as shown in FIG. Jobs that do not require any crew members j4... Jobs that do not require any crew members at all j5... Jobs that do not require any crew members at all j6... These are given as jobs that do not require any crew members at all.

On the other hand, multiple pieces of resource information registered in the resource database 2 include, for example, as shown in Fig. 3, a...people who hold regular licenses, b...people who only hold moped licenses, etc. C... Person who holds no license d... Person who holds no license e... Person who holds no license f... Person who holds no license It is given as a person who is not.

On the other hand, when the job and resources are given as shown in FIGS. 2 and 3, the constraints registered in the job constraint condition investigation unit 11 are, for example, ...The person assigned to job j1 must hold a regular driver's license or a moped license.

■...The person assigned to job j2 must hold a regular driver's license.

etc. given as

The head job retrieval unit 3 selectively retrieves jobs from the job database 1 in order starting from the first registered job under the control of an allocation status judgment/algorithm determination unit 10 to be described later.

The job retrieval unit 4 with the most restrictive conditions selectively retrieves jobs from the job database 1 under the control of the allocation status judgment/algorithm determination unit 10 in order from the job with the most restrictive conditions. .

These job retrieval units 3.4 correspond to the algorithm of the allocation determination procedure, which determines which job to proceed with allocation determination in order, and are selectively driven under the control of the allocation status judgment/algorithm determination unit 10. Ru.

Therefore, when a job consisting of the job database 1 is retrieved by the first job retrieval section J or the job retrieval section 4 with the most restrictive conditions, the resource retrieval section 5 selects the resources to be allocated to this job. This is selectively retrieved from the database 2. Then, this job information and resource information are given to the allocation decision constraint/dynamic condition inspection unit 6, and it is determined whether or not the combination of these jobs and resources is possible using the constraints as described above. The investigation is being conducted in accordance with the conditions, etc. Then, according to the investigation results, resource candidates that can be allocated to the job are determined and stored in the allocation candidate resource storage section 7.

The allocation determining unit 8 selects and determines the resource to be allocated to the job from among the resource candidates stored in the allocation candidate resource storage unit 7 in accordance with the allocation determination policy. The information decided to be allocated separately by the allocation determining unit 8 is stored in the allocation status storage unit 9, for example, in the fifth
They are stored in order as shown in the figure.

The aforementioned allocation status judgment/algorithm determination unit 10
The information stored in the allocation status storage unit 9 in this way,
That is, according to the information on the combinations of jobs and resources for which allocation has already been determined, jobs remaining in the job database 1 (jobs for which allocation has not been determined) and resources remaining in the resource database 2 (jobs for which allocation has not been determined) are The allocation determination algorithm suitable for the future allocation determination procedure is selected by referring to the undetermined resources (undetermined resources).

Further, the job constraint condition investigation 8Ii11 is activated under the control of the allocation status judgment/algorithm determination unit 10, and at that time, the job constraint investigation 8Ii11 is started to check the severity of the allocation constraint conditions for the jobs remaining in the job database 1. The allocation status judgment/algorithm determination unit 10 is configured to carry out an investigation and respond with the investigation results to the allocation status judgment/algorithm determination unit 10.

The allocation status judgment/algorithm determination unit 10 determines the allocation determination algorithm to be adopted in accordance with such investigation results. As mentioned above, there are two algorithms in place: one that proceeds with the allocation decision procedure starting from the first job, and another that proceeds with the assignment decision procedure in order starting from the job with the most restrictive conditions. The first job retrieval unit 3 or the retrieval unit 4 for the job with the most restrictive conditions is selectively activated.

Now, FIG. 6 is a diagram showing an example of a procedure for determining assignment of jobs and resources in the information processing system configured as described above.

This processing procedure is basically a continuation of the t1 assignment determination process, and if there is a job to which constraints have been added due to the assignment situation, processing is performed according to the 1 assignment determination algorithm shown in Figure 8 above. If the procedure is advanced and there is no job with the constraint condition (4 added), the processing procedure is proceeded according to the allocation determination algorithm not shown in FIG. 7 described above.

That is, here, first, it is checked whether or not there is a job (jobs whose allocation has not been decided) that is stored in the job database 1 (jobs whose allocation has not been determined). Step al).

If the existence thereof is unknown, the job constraint condition investigation section 11 described in 11.eta. is activated to first investigate whether or not a constraint condition is added to each of the above jobs (step a2). After such an investigation, we decided to re-evaluate the job
It is checked whether or not there are any jobs to which constraint conditions have been added among the jobs stored in the database 1 (step a1).

If there is a job to which a constraint condition is added, the extraction unit 4 for the job with the most severe constraint condition
is activated, and the job with the strictest constraint conditions is retrieved from the job database 1 (step a3).

Then, resources are sequentially extracted from the ship allocation resource database 2 for the job extracted from the job database 1 (also at step a5). This retrieval of resources from the resource database 2 is performed by retrieving multiple resources registered in the resource database 2 for that job in order, starting from the one registered at the beginning, each time one job is given. be exposed.

Then, for the jobs and resources selectively retrieved from each of the databases 1.2, the allocation determination constraint/dynamic condition checking unit 6 checks the combination of $ reduction conditions and dynamic conditions. Checked (step a6)
. If the combination satisfies the above-mentioned constraint conditions and dynamic conditions, the information is sequentially stored in the allocation candidate resource storage unit 7 shown in FIG. It is stored (step a7).

When the investigation of constraint conditions and dynamic conditions for all of the plurality of resources for the job is completed, the resources to be allocated to the job from among the plurality of resource candidates stored in the allocation candidate resource storage section 7 are is determined (step a8).

This resource can be determined according to the allocation policy.

Thereafter, the information on the resource for which the allocation has been decided is stored in the allocation status storage unit 9 in correspondence with the job, and the information about the resource for which the allocation has been decided is removed from each database 1.2, and the allocation for the next job is made. The determination procedure (processing procedure from step a1 to step a9) is proceeded in the same manner.

However, as the processing procedure proceeds in this manner, the pending jobs remaining in the job database 1 at the time i! Only those that do not have J conditions attached. However, it is also possible that jobs to which no constraint conditions are attached do not exist at all.

However, in this case, the first job extraction 81I3
is started, and the first job among the undetermined jobs remaining in job database 1 is retrieved (step a).
4). Then, the above-described resource candidate extraction process and the resource determination process (processing from step a5 to step a9) are similarly executed for this first job.

According to this method, in which the allocation decision procedure for jobs and resources is proceeded as described above, the algorithm for the allocation decision is 1! It will be used selectively from a number of algorithms. In the example shown in this example,
The resource allocation determination procedure is proceeded in the order of jobs having constraint conditions, and then the resource allocation determination procedure is proceeded for jobs without constraint conditions. As a result, resources are allocated with priority given to jobs with stricter constraints, so the probability that the allocation procedure stalls is extremely low.

Also, the need to change the allocation is greatly reduced.

Therefore, it is possible to reduce the processing time required for the allocation determination process.

In addition, if there are no more jobs with -1 constraints attached, the allocation is simply determined in order starting from the first job, so compared to the case where the allocation is determined while determining the constraint conditions for all jobs. , the processing procedure can be simplified and the processing burden can be reduced.

In other words, for the conventional allocation decision problem,
It was necessary to specify in advance what kind of processing algorithm to use. However, according to this method, the algorithm that is determined to be the most appropriate is selected from among multiple algorithms according to the nature of the job to be used for allocation determination, without specifying the algorithm to be adopted in advance. Then, the allocation determination process will proceed. In addition, since the status of the allocation is determined, if the algorithm being used at that time becomes inappropriate during the process of determining the allocation, the algorithm will be automatically changed according to the situation at that time. It turns out.

Therefore, it is possible to efficiently proceed with the assignment determination process while employing an appropriate assignment algorithm according to the nature of the job to which the assignment is to be determined. In addition, since the algorithm for allocation determination is selected in accordance with the situation in this manner, the occurrence of deadlocks in allocation determination can be suppressed, and the processing efficiency can be significantly improved.

Note that the present invention is not limited to the above-described embodiments; for example, the types and number of allocation determination algorithms prepared in the rP4 system may be determined according to the specifications of the allocation determination system. be. Further, although the explanation has been given here using an example of resource allocation to jobs, it is of course possible to allocate and determine jobs to specified resources as an algorithm for determining the allocation. Furthermore, it is of course possible to switch between the job-based allocation decision and the resource-based allocation decision depending on the situation.

Furthermore, the selection criteria for the allocation determination algorithm can be modified in various ways depending on the allocation policy. Furthermore, the allocation determination procedure may be carried out according to an arbitrary algorithm at the beginning, and when the procedure has been completed to a certain extent, the allocation status may be reviewed and the algorithm may be changed. In summary, the present invention can be implemented with various modifications without departing from the gist thereof.

[Effects of the Invention] As explained above, according to the present invention, a plurality of algorithms for the allocation decision procedure are prepared, and an appropriate allocation decision algorithm is selected according to the situation of the allocation decision, and the allocation is determined. The process is very efficient, even for large numbers of jobs and resources, and with little to no gridlock in the allocation process, i.e., few changes to allocation decisions. This has great practical effects, such as making it possible to determine the allocation.

FIG. 1 is a schematic configuration diagram of an information processing system configured by applying an embodiment of the present invention, FIG. 2 is a diagram showing an example of a job, FIG. 3 is a diagram showing an example of resources, and FIG. 4 is a diagram showing an example of a resource. The figure shows an example of constraint conditions, Figure 5 shows an example of the result of resource allocation to a job, Figure 6 shows the allocation decision processing procedure in the embodiment method, and Figures 7 and 8 FIG. 3 is a diagram showing a conventional general allocation determination processing procedure.

DESCRIPTION OF SYMBOLS 1...Job database, 2...Resource database, 3...First job retrieval section, 4...Job retrieval section with the most restrictive conditions, 5...Resource retrieval section, 6..・Allocation determination constraint/dynamic condition inspection unit, 7... Allocation candidate resource storage unit, 8...
Allocation determination unit, 9... Allocation status storage unit, 10...
Allocation status judgment/algorithm determination unit, 11...Job constraint condition investigation unit.

Applicant's agent: Patent attorney Takehiko Suzue Figure 1

Claims (3)

[Claims] (1) When determining the allocation of jobs and resources according to the allocation constraint conditions for the jobs and resources described above, refer to the allocation status and the information on the jobs and resources in the process of allocation determination, and An allocation decision support system characterized in that an allocation decision algorithm to be used for deciding the assignment between the job and the resource is selected from among assignment decision algorithms. (2) The allocation determination support system according to claim 1, wherein the allocation determination algorithm is selected each time in the allocation determination process. (3) The allocation determination algorithm is one that is selectively determined by referring to all or part of the allocation status and job and resource information in the allocation determination process. Decision support method.