JP2010039555A - Workflow management system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a work flow management system for achieving the sharp reduction of management man-hours or a lead time. <P>SOLUTION: A workflow management system manages the flow of an operation by using a computer. The workflow definition is provided to define the operation contents and operation procedures of a plurality of tasks, and to define an evaluation expression for evaluating the executing situations of each task. The operation content of each task is defined by using entity having variable state quantity as data necessary for executing the operation and data showing the operation result. The evaluation expression is provided to define prescribed conditions to be satisfied by the state quantity of the entity. In the case of executing the task, the evaluation value of the state quantity of the entity is calculated by using the evaluation expression, and validity/invalidity of the state quantity is decided. When the state quantity of the entity satisfies prescribed conditions, the next task is executed. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a system for managing a workflow in, for example, the manufacturing industry, and more particularly to a workflow management system that realizes a significant reduction in management man-hours and lead time.

  A workflow standardizes and describes a series of work flows. In general, a person in charge is set for each work, and a person designated as the person in charge performs the work. FIG. 14 shows an example of an article purchasing workflow in a company. The workflow management system is designed to give work instructions to the person in charge in the defined order so that work can be performed in the order defined in the workflow, and to prepare necessary electronic documents so that the work can be performed smoothly. Is. Patent Document 1 discloses a workflow management system aimed at enabling anyone to execute the same work method or start work without being aware of preparation. For each defined task, success or failure is defined, and it will be in either state at the end of the task. Also, as shown in FIG. 15, when a failure occurs, exception processing or correction processing (compensation processing) can be executed. Here, exception processing refers to special processing that is performed only when a failure occurs, and correction processing refers to restoration processing that restores the contents that have been executed.

JP 2008-65580 A

  However, it is difficult to describe the cause when the situation does not become as expected and the case where the recovery process becomes complicated, such as a defect in the manufacturing industry. For this reason, when there are many situations where it is difficult to describe, the work often cannot be completed on the workflow management system, and as a result, the work may be performed independently without using the workflow management system. In such a situation, there is a high possibility that it will not be possible to expect the “improvement of work efficiency through work identification”, which is a great merit of using workflow. In addition, since only part of the man-hour information remains, there is a possibility that the utility value may be significantly reduced as information for calculating by using the actual man-hours such as calculation of manufacturing cost.

For example, consider a workflow example in mold manufacturing as shown in FIG. Here, it is considered that a work process having the following characteristics is made into a workflow.
-Not only data but also physical objects appear in the work process, and there are various types.
-Work resources (physical objects, people such as workers) are finite (and not considered infinite), and resource contention frequently occurs between tasks.
-With deformation and change of entities that appear in the work process, such as physical deformation and chemical reaction.
-It is necessary to change the work order and work contents according to the above changes.

Many such work processes exist in the manufacturing industry and have the following characteristics.
-Although there is a series of work flows, the desired results are often not obtained even if each process is performed according to the procedure. In the manufacturing industry, even if the same procedure is used, different results are generally obtained depending on conditions such as temperature and humidity, material non-uniformity, machine differences, and the like. In such a situation, if you do not add work that is not defined in a series of procedures or change part or all of the flow to a different procedure, you will not be able to obtain the desired results, Will appear.
-Even if the work order is fixed, it is necessary to adjust the work start or change the work contents (to satisfy the delivery date) depending on the resource contention and resource status.

  For example, let us consider a case where the work process in the mold manufacturing as described above is made into a workflow. In processing in mold manufacturing, it can be determined by looking at the shape of a part whether or not various processing steps are executed in order to create a mold part. However, in reality, it is not often that processing will be completed if it is flown according to the process, and processing data with parameters changed while measuring is created and processed again, sandwiching a completely different process for adding manual processing. In some cases, it is necessary to change the work flexibly according to how the material is cut at that time, such as a process for repairing the excessively cut material, and the situation of the processing machine or tool to be actually processed.

  When such a situation occurs frequently and the cause and the recovery process are not as expected, it is difficult to describe the whole in the conventional workflow. In the real world, even if the work is completed normally, the result is not always obtained as expected, and there is little causal relationship between the success / failure of the work and the result generated as expected. For this reason, simply executing a conventional workflow management system that determines only the content of work and its success and failure will result in many situations where the work has progressed normally but the result is not desired. Therefore, the meaning of executing according to the workflow is lost.

  Furthermore, there are other situations that cannot be realized with a conventional workflow management system. That is, when the deliverable is different from the assumed situation, it is necessary to determine the corresponding process according to the deliverable state, but it cannot be described in the workflow. This is because there is no concept of a product state in the workflow. In general workflow, processing at the time of error is limited to exception processing or correction processing. However, in exception processing, the restart position after the end of processing cannot be freely selected, and correction processing can only be restored. Since it is not possible to choose, free rework cannot be realized. When describing as a flow of work rather than a state of deliverables, it is necessary to describe everything as a (pseudo) work or branch and connect the flow to the place where work is resumed depending on the situation, It will be necessary to describe m × n flows for the number of tasks in which problems occur: m and the number of tasks to be resumed: n. Increase.

  In addition, in the manufacturing industry, competition for non-human resources such as production machines, tools, and jigs frequently occurs. However, in conventional systems, it is possible to take into account such resource competition. Because it was not possible, it was determined that work was possible on the system, but there was a situation where work was not possible. As a result, it was necessary for a person to judge waiting for a resource, which did not contribute to the improvement of productivity. The system did not automatically provide a reason for not being able to start work, and therefore could not provide enough information to optimize resources and predict production in production activities in the manufacturing industry.

  In addition, the worker is defined in the workflow, and the necessary data is also defined there, but the data can be regarded as not needing to consider conflicts (data such as files and document formats, money , Only paper that can be printed as much as possible). Therefore, it is not considered in the first place to change the work order or change the work contents according to the vacant state of a finite number of items, which is important in a work process such as manufacturing industry. As a result, if work is expressed using a conventional workflow, the system may not be able to actually work even if it recognizes that the work is possible (for example, things are competing). In such a situation, the conventional workflow management system is almost powerless, and it is necessary for a person to determine whether or not the work can be performed. This has not led to an improvement in productivity.

  In view of the above-mentioned matters described above, the present invention is a resource that has a finite number of resources that are complicated and that execution according to a defined work flow does not always lead to a correct result and it is difficult to express complicated rework. The main objective is to achieve a significant reduction in management man-hours and lead time by improving the accuracy of manufacturing man-hour management and production prediction, especially in the manufacturing industry. To do.

The first aspect of the present invention is:
A workflow management system for managing a work flow using a computer,
Workflow definitions that define the work contents and work order of multiple tasks, the preconditions for executing each task, and the evaluation formula for evaluating the execution status of each task are stored as data in the memory.
The work content of each task is defined by using an entity having a variable amount of state as data necessary for executing the work and data related to the work result,
The evaluation formula defines a predetermined condition that the state quantity of the entity should satisfy,
When executing any one of the plurality of tasks, the evaluation formula is extracted from the memory to monitor whether the state quantity of the entity related to the task satisfies a predetermined condition, and the evaluation formula An evaluation formula determination means for determining an appropriate value of the state quantity by obtaining an evaluation value of the state quantity of the entity using
When the state quantity of the entity changes due to execution of a task or notification from the outside, state quantity update means for updating the state quantity of the entity to the changed state quantity;
The association between the evaluation formula and the entity is defined, the change of the state quantity of the entity related to the evaluation formula extracted by the evaluation formula determination unit is monitored, and the state quantity after the change when the state quantity changes Entity sensor means for transmitting to the evaluation formula determination means,
A workflow management system comprising: a flow control unit that accepts whether or not the state quantity of the entity is appropriate by the evaluation formula determination unit, and starts execution of a next task when the state quantity of the entity satisfies a predetermined condition It is.

  According to a second aspect of the present invention, in the workflow management system according to the first aspect, the flow control unit receives a determination that the state quantity of the entity does not satisfy a predetermined condition from the evaluation expression determination unit. In this workflow management system, the task associated with the entity to be evaluated is executed again.

According to a third aspect of the present invention, in the workflow management system according to the first aspect, when a change occurs in the state quantity of the entity, the evaluation value of the evaluation formula associated with the change in the state quantity History management means for receiving changes from the evaluation formula determination means as data together with changes in the task and the state quantity, and storing them in a memory;
When the flow control unit receives a determination from the evaluation formula determination unit that the state quantity of the entity does not satisfy a predetermined condition, the flow management unit stores the entity as data for each task by the history management unit This is a workflow management system that refers to a change in state quantity and a change in evaluation value, and designates a task in which a change occurs in the state quantity and evaluation value as a task to be re-executed.

Furthermore, a fourth aspect of the present invention is the workflow management system according to any one of the first to third aspects, wherein the task work order includes a plurality of tasks to be processed in parallel, and the task is executed. If the entities required to do conflict between tasks, a priority is assigned to each task to execute the task using the entity,
The flow control means starts task execution when the state quantity of the entity satisfies a predetermined condition and the task is in an executable state with reference to the priority order. It is a workflow management system.

According to the first to fourth aspects, in the workflow definition, in addition to the work contents and work order of the tasks defined in the normal workflow, the task work contents are defined using an entity having a variable state quantity. When the state quantity of the entity changes, it is possible to advance the flow while evaluating whether the task has achieved the desired result by evaluating the suitability of the state quantity of the entity after the change. It becomes possible.
In addition, according to the first to fourth aspects, by associating an entity and a task in the workflow definition, not only the normal flow work but also the rework work location is specified, and the rework is performed on the work flow system. Can be realized.
Furthermore, according to the first to fourth aspects, in the workflow definition, when there are a plurality of tasks to be processed in parallel and the entities required to execute the tasks compete between the tasks, By defining the priority of each task as an order, it is possible to avoid contention among a plurality of tasks processed in parallel.
As described above, according to the first to fourth aspects, in a workflow, for example, it is difficult to express a complicated rework that does not necessarily lead to a correct result when executed according to a defined work flow. It is possible to solve the conventional problems such as not considering only resources.

  According to the present invention, particularly in the manufacturing industry, it is possible to realize a significant reduction in management man-hours and lead time.

  Embodiments of the present invention will be described below with reference to the drawings. However, the following description is merely an example of the present invention, and the technical scope of the invention is not limited by the following description.

[Workflow management system configuration]
The workflow management system 100 according to the present embodiment includes a general computer terminal such as a personal computer or a workstation and its peripheral devices, and a computer program executed on the computer. Since the computer peripheral device is a general-purpose device, it is the software part related to the workflow control that plays a central role in this system. FIG. 1 shows a schematic diagram of a workflow management system 100 according to the present embodiment.

  The workflow management system 100 according to the present embodiment stores a workflow definition that defines a series of work flows as data on a computer memory, and generates a workflow using the workflow control module that is software according to the present embodiment. In addition, it is realized by controlling each task constituting the workflow.

  In addition to defining the work order of each task and the work content of each task as generally defined in the prior art, the workflow definition 110 includes an evaluation expression as a condition for starting a task and a task state as a task state. It is possible to define an evaluation formula for determining whether or not is performed as expected for each task. These evaluation formulas are stored in the computer in association with entities necessary for defining each task. The entity referred to here is defined as variable input data that is necessary to execute a task and affects the execution result, and output data after the task is executed. That is, the evaluation formula is associated with an entity necessary for evaluating whether or not the task is in an executable state and evaluating whether or not the execution result of the task is appropriate. By defining such entities in association with evaluation formulas, create a mechanism that allows a computer to quantitatively determine whether a task is in an executable state based on changes in the entity's state quantity. Can do. In addition, when a task execution result changes with a change in the entity's state quantity, create a mechanism for the computer to quantitatively determine whether the task execution result is as expected by using an evaluation formula. Can do. In particular, by defining the failure status of the task execution result that is usually difficult to judge unless it is a so-called expert by using such entities and evaluation formulas, it is possible to identify undesirable tasks before proceeding to the downstream process of the flow. The execution result can be extracted.

  In addition, when entities affect the tasks in complex relations between tasks in upstream and downstream processes such as mold manufacturing, a common entity should be defined with an evaluation formula in each task. Even after the task has progressed to the downstream process, it can be efficiently used to re-execute the task by going back to the task in the upstream process.

  The evaluation formula determination module 120 receives an evaluation formula defined in the workflow definition, extracts an entity necessary for each evaluation formula, and registers it in a memory or a database. While executing the task, the evaluation formula determination module 120 satisfies the condition by an evaluation formula that defines a predetermined condition that the state quantity of the entity 140 stored in the memory, the database, or the like should satisfy. Determine whether or not.

  The application 180 is software having a set of commands necessary for executing each task. When the application 180 is executed, the state quantity of the entity related to the task is also registered in a database or the like. As described above, the application 180 is configured such that the state quantity of the entity is constantly updated as the task is executed. Therefore, when an entity obtained as a result of execution of a task affects the execution of the subsequent task, the state quantity of this entity is updated as needed, so that the evaluation formula for the subsequent task is updated. Can be evaluated based on the latest information.

  The entity sensor module 130 is a module that monitors a change in the state quantity of the entity accompanying the execution of the application 180 and notifies the evaluation expression determination module 120 of the change when the state quantity of the entity changes.

  The flow control module 150 controls the start and end of the workflow, and when it is determined by the evaluation formula that the task does not generate a desired execution result, the task associated with the entity necessary for the evaluation formula evaluation And the execution of the task is controlled to wait for the change of the entity until the evaluation value of the evaluation formula satisfies a predetermined condition. For this purpose, when the workflow is started, the flow control module 150 reads each task from the workflow definition 110, associates the work order of each task and the work content with the entity, and then sets each task and an application that executes the task. A workflow 160 is generated as data to be included. Then, the flow control module 150 further associates the associated task and entity with the evaluation expression, and ends the task when the evaluation value of the evaluation expression obtained from the evaluation expression determination module satisfies a predetermined condition. Control to do.

  Further, the workflow control module 150 not only waits for an entity change while repeating the determination by the evaluation expression until the evaluation value of the evaluation expression satisfies a predetermined condition, but also when the evaluation value of the evaluation expression does not satisfy the predetermined condition. In order to change the state quantity of the entity related to the evaluation formula, it is also possible to control the execution of the task so as to return to the task related to the change of the state quantity of the problem entity upstream from the current task. Good. In this case, in order to change the evaluation value so as to satisfy a predetermined condition by changing the state quantity of the entity, based on the association between the task, the entity, and the evaluation formula, which task upstream of the current task In the workflow definition 110, a rework task for rework that defines whether or not should be re-executed is defined. In this way, by defining the workflow definition 110, the workflow control module 150 not only controls the start and end of each task but also the task to change the entity when the task is in an inconvenient state. It is possible to perform control so that a specific task upstream is re-executed as a return work, and then waits until the evaluation value of the evaluation formula satisfies a predetermined condition in the current task due to a change in the entity.

  When notified from the workflow control module 150 of the end of the previous task and the start of the next task, the task control module 170 activates the application 180 for executing the next task. When the task control module 170 receives a notification from the workflow control module 170 that the evaluation value of the evaluation formula does not satisfy a predetermined condition while executing a certain task, the state quantity of the entity related to the evaluation formula In order to change the application, the application 180 is started to execute an upstream task related to the entity. When the task control module 170 receives a task end notification from the application 180, the task control module 170 sends a task end notification to the workflow module 150.

  The assignment module 190 is a module that determines whether or not a task can be physically executed when the task control module 170 is notified of the task execution start from the workflow control module 150. The allocation module 190 receives from the task control module 170 an inquiry as to whether or not the task to be executed is actually executable, and transmits a result of whether or not the task is executable to the task control module 170. As a more specific example, regarding die machining, if a machine that is used in a task such as cutting is used in parallel with another task at the same time, the machine is only used for a specific task. It is necessary to decide beforehand that it cannot be used. In order not to compete between tasks that are processed in parallel, such as the above-described processing machine, it is defined in advance which task is preferentially executed when the entities compete. That is, task priorities are assigned to each of the tasks that the entity can use at the same time. By defining in this way, the task control module 170 can determine whether or not the task can actually be physically executed in a state in which the workflow control module 150 is notified of the task execution start. it can.

[Examples of workflow, task, entity and evaluation formula]
Here, workflow definitions, tasks, and entities will be described using simple specific examples. FIG. 2 shows a workflow definition and a task definition of an operation of “loading 500 to 1000 kg of luggage on the loading platform, and if the fuel is 80% or more, the truck departs from the garage”. The tasks in this case are defined as three tasks of “carrying the load”, “loading the load on the loading platform”, and “departing the truck”. In addition, the entity of truck has two state quantities, one is “loading amount of loading platform” and the other is “amount of fuel”.

  The next task after the completion of the task of “carrying the load” is “loading the load on the carrier”. The evaluation formula for evaluating the end of the task of “loading the load onto the carrier” is “loading amount of the truck carrier> 500 kg AND loading amount of the truck carrier <1000 kg”. The workflow control module 150 notifies the task control module 170 of the completion of the task of “loading the load on the bed” only when the evaluation formula is satisfied.

  On the other hand, if the evaluation formula regarding the state quantity “amount of fuel” of the entity “truck” is not satisfied at this time, the workflow control module 150 cannot notify the start of the next “truck departure” task. An evaluation formula for evaluating whether or not the task of “truck departure” can be started is “amount of truck fuel> 80%”. The workflow control module 150 controls the task start of “truck departure” when the above two evaluation formulas relating to the state quantity “loading capacity of the loading platform” and “fuel quantity” of the entity “truck” are satisfied. Notify module 170.

[Workflow control flow]
A flow for controlling a workflow using the above workflow definition will be described below with reference to FIG. The workflow control module generates a workflow and controls the start and end of execution of each task constituting the workflow.

  First, when the user is prompted on the computer and the workflow control module 150 is activated (step 300), the workflow control module accesses the workflow definition 110, reads the task defined together with the entity and the evaluation formula, and the workflow. Is generated (step 301). When the workflow definition is read, each flow is associated with an entity that affects the execution result (step 302). By steps 301 and 302, the work order of each task and the work content of each task are defined as data for the workflow, and the state quantity of each entity in each task and the evaluation formula are defined as data in association with each other. Become. The workflow control module 150 transmits a start notification to the task control module 170 (step 303).

  On the other hand, the evaluation formula determination module 120 accesses the workflow definition 110 and reads the evaluation formula defined together with the entity (step 304). Thereafter, the evaluation formula determination module 120 registers an entity sensor that monitors the state of the entity necessary for evaluating the read evaluation formula (step 305).

  The task control module 170 receives a notification of the start of the next task to be executed from the workflow control module 150 (step 310). As will be described later, at the stage where the second and subsequent tasks are to be executed, notification of the end of the previous task is received and notification that the evaluation formula related to the task satisfies a predetermined condition Is also received (step 310). Wait until all notifications to be received in step 310 have been received (step 320), and once all notifications have been received, the assignment module 190 determines whether the task being executed is actually executable. The inquiry is received from the task control module 170, and the result of whether or not it is executable is transmitted to the task control module 170 (step 330). Specifically, in order not to compete between tasks in which the entities are processed in parallel, it is defined in advance which task is preferentially executed among the tasks that are processed in parallel when the entities compete. That is, the priority for the entity is assigned to each task that can be processed in parallel. Then, the allocation module transmits a result indicating whether or not the task is executable based on the allocated priority order to the task control module. When the task control module 170 receives the result that the task is executable from the allocation module 190, the task control module 170 executes the application corresponding to the task (step 340).

  Next, the application 180 monitors whether or not the state quantity of the entity has changed due to the execution of the application corresponding to the task (step 350). When the state quantity of the entity changes due to the execution of the application, the entity state quantity after the change is registered in the database, and the entity sensor module detects the change of the entity (step 351). Note that the application is created so that the latest entity is always input to the application from the database, and when the entity after execution of the application changes, the entity is registered as update data in the database. Thereafter, the evaluation formula determination module 120 evaluates the evaluation formula using the state quantity of the entity after the change (step 352). The evaluation formula determination module 120 checks whether the evaluation value of the evaluation formula has changed (step 353). When it is determined that the evaluation value has changed and the predetermined condition is satisfied, the workflow control module 150 notifies the workflow control module 150 that the evaluation value has changed (step 354). Is notified to the task control module 170 (step 355).

  On the other hand, when the entity sensor module 130 monitors the entity state quantity and the entity state quantity does not change as a result of execution of the application, the flow control module 150 accepts that fact from the evaluation expression determination module 120 and terminates the application. Is notified to the task control module 170 (step 360), the task control module 170 notifies the workflow control module 150 of a task end notification (step 370). Thereafter, the workflow control module 150 checks whether the next task exists (step 380). If the next task does not exist, the entire flow is terminated. On the other hand, if the next task exists, the workflow control module 150 notifies the task control module of the end of the task in accordance with the workflow definition (step 381).

  As described above, in the present embodiment, for the second and subsequent tasks in the workflow, in step 310, in step 303, step 355, and step 381, the start notification of the target task to be executed is related to the target task. The target task is executed after waiting for the notification that the evaluation formula satisfies a predetermined condition and the completion notification of the task preceding the target task. Therefore, whether the task to be executed is actually executable in relation to other tasks that can be executed in parallel, or the task to be executed is executed with the state quantity of the desired entity. It is possible to automatically determine whether or not the task has been executed, and it is possible to realize a mechanism for automatically determining whether to execute the task again or terminate the task. That is, normally, when there is no problem in the result of the task, the next task is executed as it is. On the other hand, in order to avoid the conventional method from being overlooked and proceeding to the next task even though a problem has occurred, in this embodiment, a checklist for preventing the problem from occurring. Is defined as an evaluation expression associated with the task, and entities necessary for evaluating the evaluation expression are defined. And the change of the state quantity of the entity defined on the computer in this way and the change of the evaluation value of the evaluation formula accompanying the change are continuously monitored. In this respect, it is important to specify in advance which task an entity and an evaluation expression can relate to. In die machining with shape control, which will be described later, there are entities that are closely related to multiple processes (tasks), so the change in the entity's state quantity in the task being executed is the result of the upstream task execution. May be associated with not necessarily being appropriate. In such a case, a desired workflow result cannot be obtained unless a mechanism for appropriately returning the task is prepared. The fact that the change in the state quantity of the upstream task entity has an influence on the execution result of the downstream task should be taken into consideration. In this embodiment, when the entity of the downstream task changes, the execution status of the task , And return to the related upstream task from the evaluation result of the entity state quantity, change the entity state quantity, and execute the subsequent tasks again to solve the problem of the task Is possible.

[Database and entity sensor module]
FIG. 4 shows a configuration example of the entity sensor module 400 when the database 140 is used. The database access routing 410 is part of the entity sensor module 400 (corresponding to 130 in FIG. 1) and provides access routing to the database to the database access module 420 upon receipt of a modified state quantity of the entity from the application 180. At the same time, the database access module 420 is requested to update the entity state quantity. The database access module 420 inserts, deletes, and updates the database 140 as part of the entity sensor module 400. The database access module 420 accepts all processes of entity insertion, deletion, and update. Each time an update process is received, the database access module 420 notifies the evaluation formula determination module 120 of a change in the state quantity. In the case of insertion and deletion, the number of target entities will increase or decrease. It should be noted that it is possible not to use a database, and it is easily understood that such a configuration can be realized in the present embodiment.

[Return function]
As described above, when the evaluation value of the evaluation expression does not satisfy the predetermined condition, the workflow control module is configured to change the state quantity of the entity related to the evaluation expression to an entity that is a problem upstream from the current task. The execution of the task can be controlled so as to return to the task related to the change in the state quantity.

  FIG. 5 shows a flow when executing the return task. When the task control module grasps the abnormal termination of the application (step 500), the application for reversion determination is executed from the workflow definition (step 510). The system automatically recognizes the entity in which the problem occurs by the reversion determination application (step 520). This is realized by the flow control module receiving the evaluation value when there is a problem from the evaluation expression determination module recognizing the entity associated with the evaluation expression. Thereafter, the flow in which the entity is registered when the workflow is generated is searched, and a task related to the entity is extracted (step 530). When the evaluation value of an entity state quantity changes in the evaluation expression determination module, the change in the state quantity of the entity is stored in a storage means such as a memory together with the change of the evaluation value of the task and the evaluation expression. This makes it easy to extract a task related to the entity as a candidate for a task to be reworked. Thereafter, the task extracted by the search is re-executed (step 540). After re-execution, an application that returns directly to the task in which the problem occurred may be used, or an application that is re-executed from the task that is normally executed next to the task after the re-execution may be used. . In any case, a return task can be realized by accessing and executing such an application via the flow control module and the task control module.

[Evaluation expression decomposition]
FIG. 6 shows a configuration example of the evaluation formula determination module 120. When the evaluation expression definition is input from the workflow definition 110, the evaluation expression determination module 120 decomposes the evaluation expression so that the left side and the right side of the evaluation expression are unary, and further forms a tree structure. On the other hand, when an entity is input from the entity sensor module 130, the state quantity of the entity is applied to the right side of the evaluation formula, and each evaluation formula is evaluated. Thereafter, the results of the respective evaluation expressions are combined. Specifically, when the evaluation value of any evaluation expression does not satisfy a predetermined condition, the evaluation value is transmitted as data to the flow control module 150 together with that fact. Is done.

[Example of task notation]
FIG. 7 shows a task notation example (a) according to the prior art and a task notation example (b) according to the present embodiment. For example, considering the situation where there are two states of an object, which may occur suddenly in each task, and task A or B must be performed accordingly, in the prior art, FIG. Although it is necessary to describe as shown in (a), when the present technology is used, it is possible to describe as shown in FIG. 7B, and the description can be greatly simplified.

  Here, an example is shown about the workflow management system 100 concerning this embodiment.

For example, in the manufacturing process in mold manufacture, it is necessary to cope with the following situation.
-If the tool breaks, replace the tool.
-If the workpiece is scraped, replace the workpiece.
-If there is uncut material, perform additional machining.
-If there is a problem with the NC, re-create the NC.
-If the machine breaks, perform machine maintenance.
-Although everything is finished, I can't mold well.
And these situations can happen simultaneously.

  FIG. 8 conceptually shows the overall workflow of mold part creation. As a mold part creation flow, first, a mold design 800 is performed, and then NC creation 810 is performed. Then, die part processing 820 is actually performed. If a problem occurs, the process returns to the creation of the NC file 830, for example.

  Here, first, a case where a tool breakage failure occurs in a mold part machining task will be described.

  FIG. 9 shows a detailed flow of die part processing. FIG. 10 shows an embodiment 1000 of a module at the time of tool breakage as a workflow control module. FIG. 11 shows a flow that must be initially processed to enable execution of a tool change task when a tool is broken. FIG. 12 shows a flow of processing of the module when the tool is broken.

  With reference to FIG. 9 to FIG. 12, the flow of processing when tool breakage occurs will be described.

  1. First, before the tool breakage flow is executed, the evaluation formula extraction module 1010 extracts the evaluation formula definition 1020 from the workflow definition 110 (step 1100 in FIG. 11), and the tool sensor module (entity sensor module) 1050 The “breakage state” to be monitored, which is a state quantity, is read. Then, monitoring of the broken state of the tool entity 1060 in the database is started (step 1110 in FIG. 11).

  2. Also, as a premise, the flow executed due to tool breakage is set to have a higher priority than the flow executed from the start node in the workflow definition (step 1120 in FIG. 11).

  3. When tool breakage occurs in the machine during the rough machining 910 (see FIG. 9), the machine control module 1100 that has received the tool breakage error information from the machine causes the tool entity 1060 in the database. The “breakage state” is changed from “normal” to “breakage” (step 1200 in FIG. 12).

  4). The roughing 910 is temporarily suspended (execution is interrupted) when the tool is broken.

  5. When the tool sensor module 1050 monitoring the change made in 2 above recognizes the broken state of the tool as a change in the state quantity, the tool sensor module 1050 evaluates the evaluation formula corresponding to the tool break 920 (see FIG. 9). In order to execute the determination, the broken state is sent to the evaluation formula determination module 1030 as the state quantity of the tool (step 1210 in FIG. 12). The evaluation formula determination module 1030 determines the evaluation formula, changes the evaluation value of the evaluation formula, and notifies the flow control module 1070 that there is a problem of tool breakage (step 1220 in FIG. 12). When the flow control module 1070 executes the tool breakage task 920 according to the workflow definition, the tool breakage task 920 has a condition of “true if there is a tool breakage” (step 1230 in FIG. 12). Task information of the tool change task 930 (see FIG. 9) is generated according to the flow definition (step 1240 in FIG. 12).

  6). Based on the priority of the task that the tool change task has a higher priority than the roughing task, the flow control module 1070 executes the tool change 930 task by the tool change instruction application 1090 (step 1250 in FIG. 12). In other words, in this case, when a tool change task is generated, the tool change task has a higher priority than the rough machining task, so the tool assignment is changed to the tool change task. This can be realized by defining in advance in the work order of the tasks in the workflow definition that the priority order of the tool change task is higher than that of the roughing task. Although not shown in FIG. 10, a task control module and an allocation module may be arranged between the flow control module 1070 and the tool change application 1090, as in FIG. With the task control module and the assignment module arranged in this way, the tool entity assignment made to the roughing task during the roughing task execution is changed to a tool replacement task with a higher priority, and the tool change task It is possible to make it possible to start the execution of.

  7). When the tool change 930 is finished, the task control module and the assignment module are changed so that the tool entity assigned to the tool change task is assigned to the roughing task (step 1260 in FIG. 12). Since there is no task with a high value and the allocation target is restored, the interrupted roughing task 910 can be resumed (step 1270 in FIG. 12).

  Next, as another embodiment, a description will be given of the configuration of the present system when a failure occurs in the NC in a mold part machining task. FIG. 13 shows an example of the system configuration when an NC failure occurs. With reference to FIG. 13, the flow of processing when an NC failure occurs will be described.

  1. First, before this flow is executed, the evaluation expression extraction module 1010 extracts the evaluation expression definition 1020 from the workflow definition 110. Then, the NC file sensor module 1310 reads from the evaluation formula definition 1020 that the “failure state” that is the state quantity of the NC file 1330 is monitored. The NC file sensor module 1310 starts monitoring the failure state of the NC file 1330 in the database. Here, it is assumed that the possible defect states are “has a defect”, “may have a defect”, and “no defect”.

  2. Next, the failure state of the NC file 1330 in the database is changed to “May have a failure”. For example, when an abnormality is recognized by the processing machine control module 1100 during the roughing task, the state of the NC file 1330 used for the processing is changed to “May have a defect”. It is done by doing.

  3. When the NC file sensor module 1310 recognizes a change in the state of the NC file 1330 that is an entity, the NC file sensor module 1310 sends the changed state quantity of the NC file 1330 to the evaluation formula determination module 1030. The evaluation formula determination module 1030 receives an evaluation formula corresponding to the NC failure possibility 940 (see FIG. 9) from the evaluation formula definition 1020, determines an evaluation formula related to the state quantity after the change of the NC file 1330, and NC failure is possible. The flow control module 1070 is notified that the property 940 is true. The flow control module 1070 generates an NC file confirmation task according to the workflow definition 110.

  4. Based on the priority, the NC file confirmation task is executed. Note that this can be realized by predefining that the NC file confirmation task has a higher priority than the roughing task in the work order of the tasks in the workflow definition.

  5. In the NC file confirmation task, the NC file confirmation application 1320 is executed.

  6). When the NC file confirmation application 1320 determines that there is a defect in the NC file, the defect state of the NC file 1330 in the database is changed to “defective”.

  7). Here, the original roughing is to be resumed, but since the NC file is in a state of failure, it cannot be resumed due to the assignment setting and is canceled.

  8). When there is a change in the state quantity of the entity, the history management module 1080 receives the change of the state quantity from the evaluation formula determination module 1030 as needed along with the change of the evaluation value of the task and the evaluation formula, and the like. Management of storage in storage means. The flow control module 1070 makes an inquiry to the history management module 1080 in order to identify a task in which a defect has occurred in the NC file. Then, the flow control module 1070 goes back sequentially from the task in which the defect is found, refers to the change in the evaluation value of the evaluation formula stored with the task, and the NC file becomes normal (in this case, the defect is Check for tasks that have gone out of service.

  9. Since it is understood that the process in which the NC file is in a normal state from the history is an NC creation task, the NC creation task is suggested as a return destination.

  As described above, by managing the change of the state quantity of the entity together with the task and the evaluation value of the evaluation formula by the history management module, it is possible to easily specify the rework destination. In this regard, based on the association between the task, the entity, and the evaluation formula, a rework task for rework that defines which task upstream from the current task should be re-executed is defined in advance in the workflow definition. As described above, it may be set.

  Although one embodiment of the present invention and two examples thereof have been described above, the present invention is not limited to the above-described embodiment and examples, and various modifications can be made thereto. It is easy to understand. In addition, as long as they are within the scope of matters described in the respective claims and equivalent matters, they are naturally included in the technical scope of the present invention. The above-described embodiments and examples are for defects in a specific workflow, but this is merely an example, and the present invention is not limited to this specific example.

1 shows a workflow management system according to an embodiment of the present invention. The work workflow definition and task definition are shown. 6 shows a flow for managing a workflow according to an embodiment of the present invention. The example of a structure of the entity sensor module at the time of using a database is shown. The flow when executing a return task is shown. The structural example of the evaluation type | formula determination module is shown. A task notation example (a) according to the prior art and a task notation example (b) according to the present embodiment are shown. The entire workflow of mold part creation is shown conceptually. It shows a detailed flow of mold part processing. As an embodiment of the workflow control module, a tool breakage module is shown. Fig. 4 shows a flow that must be initially processed in order to be able to perform a tool change task when a tool breaks. The flow of processing of a module at the time of tool breakage is shown. An example of a system configuration when an NC failure occurs is shown. The conventional example of the workflow of goods purchase in a company is shown. A conventional example of a workflow using exception processing and correction processing is shown. The conventional example of the workflow in metal mold | die manufacture is shown.

Explanation of symbols

100 Workflow Management System 110 Workflow Definition 120 Evaluation Formula Determination Module 130 Entity Sensor Module 140 Entity 150 Flow Control Module 160 Workflow 170 Task Control Module 180 Application 190 Allocation Module

Claims (4)

A workflow management system for managing a work flow using a computer,
Workflow definitions that define the work contents and work order of multiple tasks, the preconditions for executing each task, and the evaluation formula for evaluating the execution status of each task are stored as data in the memory.
The work content of each task is defined by using an entity necessary for executing the work and an entity related to the work result and having a variable state quantity.
The evaluation formula defines a predetermined condition that the state quantity of the entity should satisfy,
When executing any one of the plurality of tasks, the evaluation formula is extracted from the memory to monitor whether the state quantity of the entity related to the task satisfies a predetermined condition, and the evaluation formula An evaluation formula determination means for determining an appropriate value of the state quantity by obtaining an evaluation value of the state quantity of the entity using
When the state quantity of the entity changes due to execution of a task or notification from the outside, state quantity update means for updating the state quantity of the entity to the changed state quantity;
The association between the evaluation formula and the entity is defined, the change of the state quantity of the entity related to the evaluation formula extracted by the evaluation formula determination unit is monitored, and the state quantity after the change when the state quantity changes Entity sensor means for transmitting to the evaluation formula determination means,
A workflow management system comprising: a flow control unit that accepts whether or not the state quantity of the entity is appropriate by the evaluation formula determination unit, and starts execution of a next task when the state quantity of the entity satisfies a predetermined condition .   The flow control unit causes the task associated with the entity to be evaluated to be executed again when the determination that the state quantity of the entity does not satisfy a predetermined condition is received from the evaluation formula determination unit. The workflow management system according to claim 1, further comprising: When a change occurs in the state quantity of the entity, a change in the evaluation value of the evaluation formula that accompanies the change in the state quantity is received as data from the evaluation formula determination unit together with a change in the task and the state quantity, and is stored in the memory. A history management means,
When the flow control unit receives a determination from the evaluation formula determination unit that the state quantity of the entity does not satisfy a predetermined condition, the flow management unit stores the entity as data for each task by the history management unit The workflow management system according to claim 1, wherein a change in the state quantity and the evaluation value is designated as a task to be re-executed with reference to the change in the state quantity and the change in the evaluation value. The work order of the tasks is a priority order for executing the tasks using the entities when there are a plurality of tasks to be processed in parallel and the entities required to execute the tasks compete between the tasks. Is assigned to each task,
The flow control means starts task execution when the state quantity of the entity satisfies a predetermined condition and the task is in an executable state with reference to the priority order. The workflow management system according to any one of the above.

Images