JPH0962733A - Flow addition device - Google Patents

Abstract

(57) [Abstract] [Problem] Flow for accurately adding a partial path that occurs due to exception processing, detecting a group of tasks that can be executed in parallel, and creating an efficient flow with a partial path added. Provide additional equipment. SOLUTION: When a start condition is added to a label from the user interface 1 and a partial path in which the execution order of the label is determined is specified by a task by specifying a start position and an end position, the start position of the partial path is a post-end condition of the task. And end position A new task created by modifying the start condition of the task is additionally stored in the execution condition storage means 52, all paths are extracted from the flow of the set of tasks, and a set of labels that configures the paths among them is extracted. Are extracted and a set of two or more paths that can be executed under the same condition is extracted, and the order of label execution is compared among the paths in the set of paths. Create a path that includes.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a flow adding device capable of efficiently adding a new work procedure generated by a change or addition of work in a flow representing a work procedure as a partial flow to an original flow, and more specifically, Specifically, for example, as in a workflow system, work procedures used by users are expressed as flows and registered in the device, and the flow is called from the device to all users who work, and the progress management of the work is performed according to the flows. The present invention relates to a flow adding device that can efficiently add a partial flow accompanying a change in a flow in a device that supports routine work.

[0002]

2. Description of the Related Art A device that supports routine work registers a flow created by a system designer, presents a work procedure according to this flow to the user, and manages the progress of the work. Therefore, it is necessary to register in the apparatus a flow designed for the work to be applied in advance. However, since the environment surrounding the user changes from the environment at the beginning of the operation of the device with the passage of time, it is necessary to deal with work procedures (exception processing) that are not registered in the device. Since the device is not supported in the exception processing, the work efficiency is deteriorated and there is a request to add the exception processing to the flow.

[0003]

In the conventional method for solving the above-mentioned problems, the system designer creates a new flow that can also support the exception handling that has occurred once, and replaces it with the previous flow. Was registered in the device. This method allows system designers to analyze exception handling,
Since the procedure for redesigning the flow is taken, there is a problem that it takes time from the occurrence of exception processing to the handling.

As a method of adding an exception process to a flow without redesigning the flow, a case in which a condition of occurrence of the exception process and a handling procedure of the exception process are described is accumulated in the apparatus, and when the exception process occurs, A method of searching for cases and eliminating exception processing according to the cases has been proposed. However, in this case, there is a problem that an inefficient work procedure is accumulated because it is a result of performing an unprecedented work.

As described above, in the conventional technique, it is not possible to satisfy both quick handling of exception handling and streamlining of the flow after handling.

The present invention has been made in view of the above,
The purpose is to add a partial path that accompanies exception processing appropriately and detect a group of tasks that can be executed in parallel to create an efficient flow with added partial paths. To provide.

[0007]

In order to achieve the above object, the present invention according to claim 1 stores a task content of each task, which is a constituent unit of a flow, in association with a label indicating each task. Means, an execution condition storage means for storing the execution start condition and the post-end condition of each task in a label indicating each task, and a task according to the execution start condition and the post-end condition stored in the execution condition storage means To create a flow, and to read the work contents of each task that composes the flow according to the flow from the task storage unit and execute the task sequentially, add a start condition to the label, and determine the label execution order. Partial path input means for inputting a partial path on the flow by designating a start position and an end position on the flow, and a post-end part of the task at the start position of the partial path. And a flow adding means for modifying the start condition of the task at the end position to create a new task from the label forming the partial path, and additionally storing the new task in the execution condition storing means, and the flow adding means. A path extracting unit that extracts all paths from a flow created from a set of tasks stored in the execution condition storing unit additionally storing the new task, and a path out of the paths extracted by the path extracting unit. A common history extracting unit that extracts a set of two or more paths that can be executed under the same condition and that has the same set of labels, and the labels of the paths in the set of paths extracted by the common history extracting unit. A parallel path creating unit for creating a path including a group of labels that can be executed in parallel from the order relationship and a parallel path created by the parallel path creating unit are configured. And summarized in that and a parallel path storage means for storing a new task and modifications tasks to the execution condition storage means.

According to the first aspect of the present invention, when a start condition is added to a label and a partial path in which the execution order of the label is determined is specified by a task by specifying a start position and an end position on the flow. , The post-end condition of the task at the start position of the partial path and the start condition of the task at the end position are modified to additionally store a new task created from the label forming the partial path in the execution condition storage means, and All paths are extracted from the flow created from the set of tasks stored in the execution condition storage means, and the sets of labels that make up the paths in the extracted paths match, so that the paths can be executed under the same conditions. Extract a set of more than one path, compare the label execution order among the paths in the extracted set of paths, and create a path including a group of labels that can be executed in parallel from the order relation. That.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a flow addition device according to an embodiment of the present invention.
The flow addition device shown in the figure is capable of adding a new work procedure, which is generated by a change or addition of work in a flow representing a work procedure, to the original flow as a partial flow.

Before explaining the configuration of the flow addition device shown in FIG. 1, the basic matters and principles used in the present invention will be described. In this flow adding device, a unit is a task for assembling a flow, and each task has a label indicating a work entity and an execution start condition and a post-end condition added thereto. Then, by registering the tasks required for the work in the device and assembling the tasks according to the two types of conditions, that is, the execution start condition and the post-end condition,
A flow that defines the label execution order is created.

The partial path added to the flow thus created is input by the user through the user interface, but is executed at the beginning of the partial path in order to determine the additional position on the flow. The start position task and the end position task to be executed last are registered in the device, and the work procedure between the start position task and the end position task is a list of labels with start conditions added in the order of execution. Partial path.

In order to additionally register the partial path configured as described above in the apparatus, the post-end condition of the start position task of the partial path and the start condition of the end position task are corrected so that the partial path other than the start position task and the end position task is corrected. A new task is created from the label, and the newly created task is additionally registered in the device.

Further, in order to detect a parallelizable path from a plurality of paths generated by the addition of such partial paths and make it into one path, and to create an efficient flow, it is registered in the device. A flow is created from a set of tasks, all paths that can be extracted from the flow are extracted, the set of labels that make up the path match, and the set of paths that can be executed under the same conditions is composed of two or more paths. A group of tasks whose order relation is not clear by extracting all the set of paths that are set, extracting the execution history of labels on each path in this set of paths, and comparing the execution history of labels between the paths. To create a single path that can be executed by the extracted task groups at the same time, and add it to the flow to create an efficient flow.

Next, the flow adding device shown in FIG. 1 will be described. The flow addition device shown in FIG. 1 includes a user interface 1 for inputting a partial path added by a user, a task information storage unit 5 for storing a file group describing a task, and information stored in the task information storage unit 5. A flow execution device 2 that creates and executes a flow based on the flow, a flow addition device 3 that additionally registers a partial path input by the user via the user interface 1, and a parallel task group from the registered task with the partial path added. The flow parallelization apparatus 4 is configured to extract and create one path that can be executed in parallel.

The task information storage unit 5 stores a group of files describing the tasks registered in the device. As a file constituting the task information storage unit 5, 5
Reference numeral 1 is a task code storage section which is made up of task action definition program codes, and 52 is an execution condition storage section which is made up of a start condition for each task and a definition file after completion.

The task code storage unit 51 stores a task action definition program in the format shown in FIG.
(1) is an identifier attached to each action, and the part (2) enclosed by "{}" is the action definition program part.

The execution condition storage unit 52 records the start condition and the post-end condition corresponding to each task in the format shown in FIG. In the item'label ', describe the label that associates the task with the action definition program, and in the item'sublabel',
Describe an identifier that identifies a task with different start conditions with the same label, describe one condition under which this task can be executed in the item'startrule ', and in the item'nextrule', after execution of this task, Specify one task group that can be executed in parallel. Multiple'stars for one task
It is possible to add'trule 'and'nextrule'.
The set of'startrule 'is the start condition, and the set of'nextrule' is the post-condition.

Hereinafter, the task name may be represented by a pair of label and sublabel. startrule describes a plurality of element conditional expressions. In the element conditional expression, the first form element conditional expression (1) that returns “true” when a specific task is in the end state,
There are two second formal element conditional expressions (2) for judging'true / doubt 'from the value obtained during the work.

Whether the start condition is true or false is one of the startrul
'true', all startrules, if e is'true '
Is'false ', it is'false', and the true / false of startrule is'true 'if all element conditional expressions are'true', and'true 'if even one element conditional expression is'false'. False.

The post-completion condition is a constraint condition that limits the tasks that can be executed after completion of the task. Any one of the nextrules is selected according to the work status and the user's selection, and a column of task names is described in the nextrule. And selected nextrule
The tasks described in are executed in parallel.

From the start condition of each task recorded in the execution condition storage unit 52, the label execution order is determined and a flow can be created. Hereinafter, for convenience, description will be made using the flowchart shown in FIG. start is the initial state before starting work, en
d is the finished state after the work, (no.1,1) is the label
Indicates that the task of No. 1 and sublabel is 1. Concurrency executes all tasks indicated by arrows in parallel,
In branching, one of the tasks indicated by the arrow is selected and executed.

The flow execution device 2 has an execution condition storage unit 52.
This device manages the progress of work according to the information of. As functions constituting the flow execution device, reference numeral 22 in FIG. 1 is a task execution start determination unit, 23 is a task execution unit that executes a task, and 24 is a state management unit.

The task execution start determination unit 22 calls the current state from the state management unit 24, and extracts all the tasks for which the start conditions of all the constituent tasks are "true" from the set of next executable task groups. Then, the extracted task group is set as an executable task group and is transmitted to the task execution unit 23.

The operation of the task execution unit 23 is such that the user is made to select one from the set of executable task groups from the task execution start determination unit 22, and for each task of the selected task group,
Declare the start to the state management unit 24, call the operation definition program corresponding to the task from the task code storage unit 51,
The program is executed, and at the end of execution, the execution result is notified to the state management unit 24.

The state management unit 24 grasps and manages the task execution state, the task execution result, and the next executable task group from the information from the task execution unit 23 and the information from the task execution start determination unit 22. The task execution state has an initial state, an execution state, and an end state. For the task in the execution state, the post-end condition is extracted from the execution condition storage unit 52 and recorded as the next executable task group.

The operation of the flow execution device 2 will be described with reference to FIG. When the user issues a work start command (step S
51), the state management data managed by the state management unit 24 is initialized, the task execution start determination unit 22 is activated (steps S52 and S53), and when there is no executable task group (step S54), the flow chart of FIG. When the executable task group exists in the state of (b) (step S5)
4) The task execution unit 23 is activated, the execution result is registered in the state management unit 24 (step S55), and the state returns to the state of (a) in the flowchart. Next to the state of (b) in the flow chart,
When the information of the state management unit 24 is called and there is no task in the execution state (step S56), the work is finished,
When there is a task in the execution state (step S56),
The information in the state management unit 24 is periodically checked, and even if one task in the execution state shifts to the execution end state, a task that shifts to the execution end state appears. It returns to the state (step S57).

The flow adding device 3 is one part of the core of the present invention, and has a function of adding the path input to the device to the registration flow. This provides a means for quickly adding the function of the flow. The configuration of the device includes a data conversion unit 31 and a flow addition unit 32.

The data conversion unit 31 converts the partial path of the format shown in FIG. 6A input through the user interface 1 into the data of the format shown in FIG. 6B.
(1) start-task is a task that indicates the start position of the partial path, creates a post-end condition of the same format as the post-end condition recorded in the execution condition storage unit 52, and (2) to (4) label is Create a new task by allocating a sublabel that is not used in the registered task and creating a start condition and a post-end condition that are recorded in the execution condition storage unit 52 in the same description format,
(5) end-task is a task that indicates the end position of the partial path,
A start condition having the same description format as the start condition recorded in the execution condition storage unit 52 is created.

The flow adding unit 32 uses start-task and end-ta.
After determining the validity of sk, the post-end condition created in the data conversion unit 31 is added to the post-end condition of the start-task registered in the execution condition storage unit 52, and the data is added to the start condition of the end-task. The start condition created by the conversion unit 31 is added, and other new tasks created by the data conversion unit 31 are additionally registered in the execution condition storage unit 52.

The validity of start-task and end-task is corrected by the correction of start-task and end-task registered in the execution condition storage unit 52 and created by the data conversion unit 31 when the following items are satisfied. Cancel the registration of the new task.

(1) The start-task is not registered in the execution condition storage section 52. (2) If the end-task is not in the work end state, it is not registered in the execution condition storage unit 52. (3) Assembling the flow from the registered task and following the flow from the start-task in sequence, parallel branch appears, as shown in Fig. 8 (1), but the parallel merge joins until the end-task appears. Does not appear. (4) When a flow is assembled from registered tasks and the flow is sequentially followed from start-task, parallel merging appears, but the corresponding parallel branch that should appear before it does not appear, as shown in Fig. 8 (2). .

In the present invention, by inputting a partial path to the flow addition device 3, the flow execution device 2 is
The flow created from the registration task is registered as a flow in which the partial paths branch from the flow of FIG. 4 and join, as shown in the flow chart of FIG. 7.

The flow parallelization device 4 is another core of the present invention. It creates a flow from registered tasks, extracts a parallel task group from the flow, and modifies the description in the execution condition storage unit 52. , Has a function of enabling the extracted task groups to execute in parallel. The configuration of the device is shown in FIG. Four
Reference numeral 1 is a path extraction unit, 43 is a common history extraction unit, 44 is a parallel path creation unit, and 45 is a parallel path registration unit.

The path extracting unit 41 has a function of extracting a set of paths to be parallelized from the registration flow. The operation of the path extraction unit 41 will be described with reference to the flowchart of FIG.
First, a flow is created from the information in the execution condition storage unit 52 (step S11), all the paths of the flow are extracted (step S12), and a set of labels forming each path is extracted (step S13). Next, the paths having the same set of labels are set as a group (step S14), and only the group having two or more paths is subjected to the processing after (a) (step S15).

The processing after (a) is performed for each group. First, for each path in the group, the product of the element conditions of the format (2) in FIG. 3 of the start conditions of all the tasks that form the path is calculated, and this is set as the'path condition '(step S16). . Next, a subset of two or more paths is created from the paths in the group (step S17), the product of the path conditions of all the paths in the subset is calculated, and a subset that does not become false is extracted (step S18). ), And informs the common history extraction unit 43 and the parallel path creation unit 44 of the information of the paths forming the extracted subset (step S19).

The common history extracting section 44 has a function of extracting a common execution history for each label from the subset of paths transmitted from the path extracting section 41. From now on,

## EQU1 ## Let P = {p ₁ , p ₂ , ..., p _M } (M is an integer of 2 or more), and let the set of labels appearing in the path of the subset be

L = {l ₁ , l ₂ , ..., L _N } (N is an integer greater than or equal to 1), an arbitrary label l _i (1 ≦ i ≦ N), and an arbitrary path p
_The following operation is performed for _j (1 ≦ j ≦ M). Path p _j
Of the tasks constituting the task t with the label l _i
(P _j, l _i) extracting, when running path p _j, extracted task t (p _j, l _i) set Ht (p _j, l _i) of the tasks to be completed before the execution of a Ht A set of labels of each task belonging to (p _j , l _i ) is extracted, and this is set as a path p _j.
Let the execution history h (p _j , l _i ) of the label l _i above. Next, the label l _i (1 ≦ i ≦ N) is fixed, and an arbitrary p
The execution history common part of the label l _i on _j (1 ≦ j ≦ M) is extracted and used as the common history H (l _i ) of the label l _i in the path subset, and H (l _i ) is parallelized. It is transmitted to the path creating unit 44.

The parallel path creating unit 44 is the common history extracting unit 4
3 has a function of extracting the parallelism of the labels that does not exist in the subset of the paths transmitted from the path extraction unit 41 from the common history transmitted from No. 3, and creating one path. The function of the parallel path creation unit 44 will be described below with reference to FIG.

The secure each label _{l i (1 ≦ i ≦ N} ), and different tasks t a set of _{(p j, l i) (} 1 ≦ j ≦ M) and _{T (l i), T (} l i ) Consists of one task (steps S21 and S22), this task is t
If (l _i ) is set (step S23), the operation after (b) is proceeded to, and when there are a plurality of elements of T (l _i ), the sub-label s not used in the registered task having the label l _i
Extracting, create a new task _{t (l i) = (l} i, s) ( step S24), and the first subject to a termination of all tasks included in the common history H (l _i) of l _i A formal element conditional expression, a logical product of the second formal element conditional expressions of all tasks included in T (l _i ) is the second formal element conditional expression star
Create a trule and call it one star of task t (l _i ).
Set trule and proceed to the operation after (b) (step S2)
5).

The operations from (b) onward are for all labels l.
After the above operation is completed for _i (1 ≦ i ≦ N), the task t
Determine the nextrule of (l _i ) (1 ≦ i ≦ N). First,
A task set Ln _i = {} is created for all t (l _i ) (1 ≦ q _i ≦ N) (step S26), and a certain t
When the label l _i is included in the common history H (l _s ) of (l _s ) (1 ≦ s ≦ N) (step S27), t (l _s ) is added to the set Ln _i (step S28). The set Ln _i created in step 1 is set as a new endrule of task t (l _i ) (steps S29 and S30), and information is sent to the parallel path registration unit 45 (step S31).

The function of the parallel path registration unit 45 will be described with reference to FIG. 12. A path is created from the new task and the modified task created by the parallel path creation unit 44 (step S35).
When the path is presented to the user (step S36), the new condition is registered in the execution condition storage unit 52 and the correction permission of the correction task is asked to the user, and the user's permission is obtained (step S37), the execution condition storage unit is obtained. In the case of a modified task, a newly created endrule is added to 52, and in the case of a new task, the task itself is registered (step S38).

In the present invention, by activating the flow parallelization device 4, a label group having an uncertain order relation in the flow created from the registered task is extracted, and a path in which the extracted label group can be executed in parallel is extracted. It is possible to create and register.

Next, as a specific example of this embodiment, a case where this flow adding device is applied to a work progress management device for contract slip processing will be described.

FIG. 13 is a diagram showing a flow of slip processing work for failure repair registered in this apparatus, and FIG. 14 shows a partial flow to be added to the registration flow of FIG. 13 in case of urgent repair. FIG. 15 is a diagram showing a flow generated as a result of inputting the partial path of FIG. 14 to the flow addition device 3 shown in FIG. 1 and registered in the device, and FIG.
FIG. 16 is a diagram showing a flow generated as a result of parallelizing the flow of FIG. 15 by the user inputting a flow parallelization instruction to the flow parallelization device 4 shown in FIG. 1.

13 to 16, first, as can be seen from FIG. 13, before starting the operation of this apparatus, for example, as shown in FIG.
Predictable information such as 4 in the case of urgent repair is analyzed, a simple flow as shown in FIG. 13 is prepared without designing a complicated flow, and exception processing that occurs after the start of operation of the device, etc. Each time, the partial flow for exception processing is shown in FIG.
By inputting this partial flow into the apparatus, the flow grows to be manageable in many situations as shown in FIG.

Further, by creating a partial flow as shown in FIG. 14 from one user's experience of exception processing and inputting it into the device, all users who use the device can use the flow of FIG. It is possible to share the experience of exception handling in FIG.

Further, by the flow parallelization device 4, FIG.
16 is a parallelized flow as shown in FIG. 16, so that “a repair request to a contractor” and “a witness of repair work” are left to another worker, and “form a repair slip” and “approve”. It is possible to simultaneously execute “approval by the person”, and it is possible to reduce the amount of work per person and the time required for all work.

[0047]

As described above, according to the present invention,
When a start condition is added to a label and a partial path in which the execution order of the labels is determined is specified by a task by specifying the start position and end position in the flow, the post-condition and end of the task at the start position of the partial path Modify the start condition of the position task,
A new task created from a label that constitutes a partial path is additionally stored in the execution condition storage means, and all paths are extracted from the flow created from the set of tasks stored in this execution condition storage means, and this extraction is performed. Of the extracted paths, the set of labels that make up the path match, and a set of two or more paths that can be executed under the same conditions is extracted. In this set of extracted paths, the label execution order is set between the paths. By comparing and creating a path including a group of labels that can be executed in parallel from the order relation, there is no need to newly redesign the entire flow when exception processing or the like occurs, Partial paths due to exception processing and the like can be easily added, and one flow can be parallelized to improve efficiency.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a flow addition device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a data description format stored in a task code storage unit used in the flow addition device of FIG.

3 is a diagram showing a data description format stored in an execution condition storage unit used in the flow addition device of FIG.

4 is a diagram showing a flow created from a set of tasks described in the execution condition description format of FIG.

5 is a flowchart showing an operation of a flow execution device used in the flow addition device of FIG. 1. FIG.

FIG. 6 is a diagram showing a conversion example for indicating a data conversion method in a data conversion unit used in the flow addition device of FIG.

7 is a diagram showing a flow created by adding the partial path shown in FIG. 6 to the flow shown in FIG. 4;

8 is a diagram showing partial paths that are prohibited from being registered in the execution condition storage unit of FIG. 1 in the flow addition unit used in the flow addition device of FIG.

9 is a block diagram showing a configuration of a flow parallelization device used in the flow addition device of FIG. 1. FIG.

10 is a flowchart showing the operation of a path extraction unit used in the flow parallelization apparatus of FIG.

11 is a flowchart showing an operation of a parallel path creation unit used in the flow parallelization apparatus of FIG.

12 is a flowchart showing the operation of a parallel path registration unit used in the flow parallelization apparatus of FIG.

FIG. 13 is a diagram showing a flow of slip processing work for failure repair when the flow addition device of FIG. 1 is applied to a work progress management device for contract slip processing.

FIG. 14 is a diagram showing a partial flow to be added to the registration flow of FIG. 13 for the case of urgent repair.

15 is a diagram showing a flow generated as a result of inputting the partial path of FIG. 14 to the flow adding device shown in FIG. 1 and registered in the device.

16 is a diagram showing a flow generated as a result of parallelizing the flow of FIG. 15 by the user inputting a flow parallelization instruction to the flow parallelization device shown in FIG. 1.

[Explanation of symbols]

 1 user interface 2 flow execution device 3 flow addition device 4 flow parallelization device 5 task information storage unit

Claims (1)

[Claims] 1. A task storage means for storing the work content of each task, which is a constituent unit of a flow, in association with a label indicating each task, and a execution start condition and an after-end condition for each task on the label indicating each task. Execution condition storage means to be additionally stored and tasks are assembled according to the execution start condition and the post-end condition stored in the execution condition storage means to create a flow, and the work content of each task constituting the flow according to the flow And a task execution means for sequentially reading out from the task storage means for execution, and a start condition is added to the label, and a partial path in which the label execution order is determined is input by designating the start position and end position on the flow by the task. A partial path input means for modifying the post-end condition of the task at the start position of the partial path and the start condition of the task at the end position to construct a partial path. Flow adding means for creating a new task from the task and additionally storing the new task in the execution condition storing means, and a task stored in the execution condition storing means additionally storing the new task by the flow adding means. Two paths that can be executed under the same conditions, in which the path extraction means for extracting all paths from the flow created from the set of the above and the set of labels constituting the paths among the paths extracted by the path extraction means match. The common history extracting means for extracting the above set of paths and the execution order of the labels between the paths in the set of paths extracted by the common history extracting means are compared, and one label that can be executed in parallel from the order relationship is compared. A parallel path creating means for creating a path including a group, and a parallel task for storing in the execution condition storing means the new task and the modification task which compose the parallel path created by the parallel path creating means. Flow addition device, characterized in that it comprises a scan storage unit.