JP3557717B2 - Design support method and device - Google Patents

Description

[0001]
[Industrial applications]
The present invention relates to a design support method and a design support device for a product used in a company, a factory, and the like. In particular, in a product design in which design is divided and shared among a plurality of design processes, design efficiency can be improved. The present invention relates to a design support method and a design support device.
[0002]
[Prior art]
PART / CPM is well known as a typical method for managing a project such as a product design which is divided and shared among a plurality of design processes. CRT (Cropical Path Method), developed by Dr. John Presper Macheley of the University of Pennsylvania in 1959, and PERT (EgoltProv), developed by William Frazer, a consultant of the Polaris Submarine Project, developed by William Frazer. and Review Technique) regards a project as a set of tasks, finds a critical path that has an effect on the schedule of the entire project from the relationship between the predecessor relation of each task, required man-hours, and resources, and manages this critical path with priority. In this way, the project can be advanced without delaying the development period.
[0003]
On the other hand, a method has been proposed in which a model of a design object is shared by a plurality of departments to support cooperation between processes.
[0004]
Dennis Bahler, Jim Bowen, et al., ASME DE (American Society of Medicine, D.Eng.), Entitled "Constrained Networks For Current Engineering". vol. 29. P. In the article 59-67, a product is modeled by attributes that specify a design object (product) and constraints existing between the attributes, and constraints are activated based on the values of the attributes input by the designer. A system for obtaining design proposals, simultaneously examining specifications from multiple design departments and multiple viewpoints, and promptly communicating specification changes to related departments by constraint propagation that sequentially determines the value of undetermined attributes. is suggesting.
[0005]
[Problems to be solved by the invention]
The above prior art has the following problems.
[0006]
In PART / CPM, it is necessary to create a preceding relational model of a task constituting a project according to a product to be developed. In addition, in order to be able to manage a project accurately, detailed task extraction is required. Therefore, it takes a lot of man-hours to build a model for project management.
[0007]
In addition, PART / CPM cannot deal with a cycle in which tasks are preceded, such as a specification change or a partial design redo due to trial and error frequently repeated in a product design stage. For this reason, if there are multiple tasks that can be started at the same time, the task to be performed first (a serious task related to the occurrence of a cycle) is postponed, and a problem is discovered later, This caused a time loss due to the spread of problems up to the design of the product, and a delay in specifications in related departments.
[0008]
On the other hand, a method of modeling a design object using a constraint network and sharing the model in a design from a plurality of departments and a plurality of viewpoints will be considered. The constraint network can be regarded as a design procedure model that determines design attributes according to constraints, and can express detailed task precedence. However, since the constraints dealt with here can only be defined as arithmetic expressions, at present, they cannot be applied to design problems in which there are many parts that cannot be formulated.
[0009]
Also, this mechanism is not linked to scheduled project management.
[0010]
A first object of the present invention is to solve the above-mentioned drawbacks of the prior art, quickly generate a model for project management based on a model of a product design object, and use the model to execute a design process or a design process. An object of the present invention is to provide a design method capable of improving the cooperation efficiency between departments.
[0011]
A second object of the present invention is to provide a design support device suitable for implementing such a design method.
[0012]
[Means for Solving the Problems]
In order to achieve the above object, the present invention provides a method for storing individual tasks for each design process in a state in which constraints expressed as relational expressions or activities between input attributes, output attributes and inputs and outputs, and required man-hours are stored in advance. A first step of reading constraints related to the design process of each of the above, and generating a constraint graph through the entire design process by associating input / output attributes of the individual constraints;
A critical path calculation step of calculating a critical path in the constraint graph from the constraint graph and the required constraint man-hours;
A cycle detection step of detecting whether or not a cycle exists in the constraint graph;
In the above constraint graph, if there are a plurality of constraints that can be activated at the same time, raise the priority of the constraint on the path related to the critical path, or if a cycle is detected in the cycle detection Is a constraint priority determining step of performing a process of raising the priority of the constraint on the cycle,
In each design process, a step of inputting / outputting an attribute value, an executable constraint extracting step of determining which constraint can be executed from the set attribute,
In the extraction of the executable constraint, when a plurality of executable constraints are detected, a constraint priority order determining step of determining a constraint priority order according to the priority determined by the constraint priority determining step,
The constraint determined by the constraint priority order determination is read, and if the constraint is a relational expression (arithmetic expression), the output attribute value is calculated. If the constraint is an activity, the activity name is specified as a work instruction item by the designer. A constraint propagation / work order step to be displayed for
As a result of the constraint propagation / work instruction, a contradiction detecting step for checking whether or not the attribute values obtained are contradictory is included.
The coordination efficiency between the design processes is improved by grasping the precedence of the tasks over a plurality of design processes and executing the tasks of the individual design processes based on the priority order determined from the viewpoint of reducing the total design time.
[0013]
As a configuration of the design support apparatus that executes these operations, a design constraint storage means for storing individual tasks for each design process as constraints expressed as a relational expression between input attributes, output attributes and input / output, or activities. When,
Constraint required man-hour storage means for storing required man-hours when each constraint is executed;
Constraint graph expanding means for reading constraints related to a plurality of design processes from the design constraint storage means, and generating a constraint graph throughout the entire design process by associating input / output attributes of the individual constraints,
Critical path calculation means for calculating a critical path in the constraint graph from the constraint graph and the required constraint man-hours;
A cycle detecting means for detecting whether or not a cycle exists in the constraint graph; and a constraint on a path related to a critical path when there are a plurality of constraints which can be activated simultaneously in the constraint graph. Or, if a cycle is detected by the cycle detecting means, a constraint priority determining means for increasing the priority of the constraint on the cycle,
Means for inputting and outputting attribute values in each design process; executable constraint extracting means for determining which constraints in the design constraint storage means are executable from the set attributes;
In the executable constraint extraction unit, when a plurality of executable constraints are detected, a constraint priority order determination unit that determines a constraint priority order according to the priority determined by the constraint priority determination unit,
The constraint determined by the constraint priority order determination means is read, and if the constraint is a relational expression (arithmetic expression), an output attribute value is calculated. If the constraint is an activity,
A constraint propagation / work instruction means for displaying an activity name to a designer as a work instruction item,
A contradiction detecting means for checking whether or not the attribute values obtained as a result of the constraint propagation / work instruction are contradictory. Furthermore, an EOA (engineering OA) system and a remote communication system configuration using a LAN configuration and a communication network system are also possible, and the following configurations are possible.
[0014]
A workstation or personal computer connected via communication control means corresponding to each design process, a signal transmission path connecting the workstations, and at least one central workstation or personal computer connected to the signal transmission path; Each workstation for each of the design processes has at least means for inputting and outputting the attribute values, design constraint storage means, executable constraint extraction means, constraint priority order determination means, constraint propagation / work instruction means, and inconsistency detection means. The central workstation has at least the constraint graph expanding means, critical path calculating means, cycle detecting means, and constraint priority determining means.
[0015]
[Action]
The above-mentioned design method is roughly divided into two stages: a planning stage in which the entire project is planned before each design process is executed, and a stage in which the tasks of each design process are executed based on the plan.
[0016]
First, the operation at the planning stage will be described. The constraint graph developing means reads, from the design constraint storage means, constraint information of a design process related to product development based on a command for planning, which is input by the operator. Constraint information indicates the relationship between input attributes and output attributes, such as formulas, tables, and other arithmetic expressions, or activities to be performed in creating output information (eg, referencing parts catalogs, calculating strength, designing calculation formulas). Reference, operation of CAD to create a part drawing, etc.). Next, the constraint graph developing means creates a constraint graph of the entire design process based on the input / output attribute name key of the read constraint. The constraint graph expresses the precedence of tasks throughout the entire design process. Next, the critical path calculation means is activated, reads the required time of each constraint from the required constraint man-hour storage means, and specifies a critical path in the constraint graph based on the time data. Subsequently, the cycle detecting means is activated, and checks whether there is a place where a cycle may occur in the constraint graph. The location where there is a possibility of occurrence of a cycle means that, for example, the attribute value provisionally determined in the higher-level design process is significantly different from the precise attribute value obtained in the downstream design process, and is accordingly determined based on the provisionally determined value. And all specifications must be changed.
[0017]
Next, the constraint priority determining means sets importance data for each constraint. Here, the priority of a constraint belonging to a path related to the critical path (a constraint group that becomes an input of a constraint belonging to the critical path) is raised. Further, when the possibility of occurrence of a cycle is detected by the cycle detection means, the operator is inquired whether to raise the importance of the constraint on the cycle, and the priority of the corresponding constraint is determined according to the answer. increase. By registering the importance data set for each constraint as described above in the design constraint storage means for each design process, the planning step ends.
[0018]
Next, the operation at the stage of executing the task of each design process will be described. Based on the attribute information determined in the higher-level design process, the designer, if there is an attribute to be input, performs the input using the attribute input means, and starts the design. First, the executable constraint extraction unit determines which constraint in the design constraint storage unit is executable from the set attributes, and extracts a startable constraint. When a plurality of executable constraints are detected, the constraint priority order determination unit is activated, compares the importance data of the constraints, and determines the one having the greater importance as the constraint to be activated. Next, the constraint is activated to calculate the value of the output attribute. When the content of the constraint is an arithmetic expression, the local constraint propagation means is activated and automatically calculates the output attribute according to the arithmetic expression. If the content of the constraint is an activity, the work instructing means is activated, and the activity to be performed in generating the output attribute is displayed to the designer as a work instruction item (eg, refer to the parts catalog, 2: Calculate strength; 3: Refer to design calculation formula; 4: Operate CAD to create component drawing). The designer performs a design operation according to the instruction and sets an output attribute value. The inconsistency detecting means is activated based on the set attribute values, and checks whether the attribute values are inconsistent. In the same manner, each means is activated until there is no undetermined attribute, and the corresponding design process is completed. Further, in this process, the attribute value is immediately transmitted to the related process when the attribute value related to the other design process is obtained.
[0019]
【Example】
An embodiment of the present invention will be described below with reference to FIGS. FIGS. 2 and 1 are diagrams respectively showing an example of a device configuration having the design method of the present invention and an example of a functional configuration such as software. In FIG. 2, the multi-bus 11 under the control of the bus control unit 12 is connected to a disk unit 18 controlled by a disk control unit 17 in addition to the central processing unit 13. A main storage device 14, a display device 16 and a keyboard 15 are accommodated. As a result, the data input from the keyboard 15 is stored in the main storage device 14 by the central processing unit 13 and simultaneously displayed on the display device 16 or the data on the main storage device 14 is Data transfer is performed, for example, data is stored in the disk device 17 via the multi-bus 11 and the disk controller 17.
[0020]
Further, as a network system in which a plurality of workstations (or personal computers) of FIG. 2 are connected, a bus, a bus control unit, a central processing unit, a disk control unit, a main storage unit, a display device, a keyboard, a disk, a communication control unit are also provided as described above. , And a plurality of workstations including a modem are connected via the modem. FIG. 3 shows the configuration. As shown by WS1-6 (301-306), each of the workstations 1-6 has six workstations in this example, and the LAN network 310 shows a line on a loop. .
[0021]
Further, in the case of the distributed type, each of the WS 1 to 6 in FIG. 3 may have the same configuration, but in the case of the centralized type, there is at least a central workstation, in which case the central workstation is a project in which a plurality of design processes are integrated. The overall planning is delegated so that other workstations perform the tasks of each design process to achieve the design.
[0022]
As shown in FIG. 1, the functional configuration of the present invention comprises a data input unit 22, a data output unit 23, and a project management unit 24 under a planning control unit 21 at a central workstation. The project management unit 24 includes a constraint graph developing unit 25, a critical path calculating unit 27, a cycle detecting unit 27, and a constraint priority determining unit 28. Further, a workstation corresponding to each design process includes a data input unit 32, a data output unit 33, a design procedure guiding unit 34, a constraint propagation / work instruction unit 37, and a design constraint storage unit under the process support control unit 31. The design procedure guiding unit 34 includes an executable constraint extracting unit 35 and a constraint priority order determining unit 36. The constraint propagation / work instructing unit 37 includes a local constraint propagation means 38, a work instructing means 39, and a contradiction detecting means 40.
[0023]
In this regard, the processing steps will be described more specifically using the flowchart of FIG. First, the planning stage will be described. The planning control unit 21 activates the constraint graph developing unit 25 based on a command for performing a planning input from the operator using the data input unit 22 (400). The constraint graph developing means 25 reads the constraint information of the design process related to product development from the design constraint storage means 41 (401). Constraint information indicates the relationship between input attributes and output attributes, such as formulas, tables, and other arithmetic expressions, or activities to be performed in creating output information (eg, referencing parts catalogs, calculating strength, designing calculation formulas). Reference, operation of CAD to create a part drawing, etc.). Next, the constraint graph developing means 25 creates a constraint graph of the entire design process using the input / output attribute name key of the read constraint (402). The constraint graph expresses the precedence of tasks throughout the entire design process. Next, the critical path calculation means 26 is activated, reads the required time of each constraint from the required constraint man-hour storage means 42, and specifies the critical path in the constraint graph based on the time data (403). Subsequently, the cycle detection means 27 is activated, and checks whether there is a possibility of occurrence of a cycle in the constraint graph (404). The location where there is a possibility of occurrence of a cycle means that, for example, the attribute value provisionally determined in the higher-level design process is significantly different from the precise attribute value obtained in the downstream design process, and is accordingly determined based on the provisionally determined value. And all specifications must be changed. Next, the constraint priority determining means 28 sets importance data for each constraint (405). Here, the priority of a constraint belonging to a path related to the critical path (a constraint group that becomes an input of a constraint belonging to the critical path) is raised. When the possibility of occurrence of a cycle is detected in the cycle detection (404), the importance of the constraint on the cycle is notified to the operator via the planning control unit 21 and the data output unit 23. An inquiry is made as to whether or not to raise the priority, and the priority of the corresponding constraint is raised in accordance with the value of the answer input by the operator via the data input means 22. By registering the importance data for each constraint set as described above in the design constraint storage means 41 for each design process, the planning step is completed (406).
[0024]
Next, the stage of executing the task of each design process will be described. If there is an attribute to be input based on the attribute information determined in the higher-level design process, the designer performs the input using the attribute input means 32 and starts the design (410). First, the process support control unit 31 activates the executable constraint extraction unit 35, determines which constraint in the design constraint storage unit 41 is executable from the set attributes, and extracts the executable constraint (420 to 420). 430). When a plurality of executable constraints are detected, the constraint priority order judging means 36 starts up, compares the importance data of the constraints (440), and determines the one with the higher importance as the constraint to be activated (450). ). Next, the process support control unit 31 sends an instruction to the constraint propagation / work instruction unit 37 to activate the constraint and calculate the value of the output attribute. If the content of the constraint is an arithmetic expression, the constraint propagation / work instructing section 37 activates the local constraint propagation means 38 and automatically calculates an output attribute according to the arithmetic expression (480). If the content of the constraint is an activity, the work instructing means 39 is activated, and the activity to be performed in generating the output attribute is displayed to the designer as a work instruction item (490) (Example: Part catalogue) 2: Strength calculation 3: Design calculation formula 4: Manipulate CAD to create parts drawing etc.). The designer performs a design operation according to the instruction and sets an output attribute value (500). The inconsistency detecting means 40 is activated based on the set attribute values, and checks whether the attribute values are inconsistent (510). If inconsistency is detected, the process support control unit displays a warning to the operator via the data output unit 33 (530). In the same manner, each unit is activated until there are no more undetermined attributes (520), and the corresponding design process ends. In addition, when the attribute value related to the other design process is obtained in this process, the process support control unit 31 transmits the attribute value to the related process.
[0025]
An example in which the design support method according to the present invention is applied to a design project of an electromechanical product X including four processes of a plan design (P1), a mechanical design (p2), an electrical design (P3), and a product prototype (p4). To explain. FIG. 5 shows the configuration of the design support apparatus according to the present embodiment. A planning workstation 501, a planning design workstation 502, a disk 503 for storing planning design constraint knowledge, a machine design workstation 504, a disk 505 for storing machine design constraint knowledge, an electrical design workstation 506, and It is assumed that a disk 507 for storing constraint knowledge for electrical design, a product trial workstation 508, and a disk 509 for storing constraint knowledge for planning design are connected to each other via a network, and disks 503, 505, 507 for each design process. In 509, it is assumed that the constraint knowledge shown in FIG. For example, the mechanical design constraint knowledge C3 is an arithmetic expression that inputs the attribute B and outputs the attribute D and the attribute E. The constraint that obtains the values of D and E by executing the operations f3 and f4 based on the value of B The required time is one hour. C5 is an activity in which the attribute E is input and the attribute H is output, and the design operation of ACT1 and ACT2 is executed based on the value of E (for example, ACT1: refer to the parts catalog, ACT2: calculate the strength). , The required time is 5 hours.
[0026]
The case where the design support method of the present invention is implemented on the premise as described above will be described below. First, the project manager inputs an instruction to perform planning for the product X using the data input unit 22 of the planning workstation 501. The planning control unit 21 activates the constraint graph developing unit 25 (400). The constraint graph developing means 25 stores the constraint information of the design processes related to product development, that is, the plan design 502, the mechanical design 504, the electrical design 506, and the product prototype 508, into the design constraint storage means 503, 505, 507 of each design process. Read from 509 (401). Next, the constraint graph developing means 25 of the planning workstation 501 creates a constraint graph of the entire design process using the input / output attribute name key of the read constraint (402). FIG. 7 shows the developed constraint graph. Next, the critical path calculation means 26 is activated, reads the required time of each constraint from the required constraint man-hour storage means 42, and specifies the critical path in the constraint graph based on the time data (403). When the start time of C1 is set to time 0 and the possible start time of each constraint is calculated based on the constraint precedence relation and the required time on the graph, the path of C1, C2, C6, C7, and C9 shown in FIG. Is extracted as Subsequently, the cycle detection means 27 is activated, and checks whether there is a possibility of occurrence of a cycle in the constraint graph (404). In the constraint C10, the attribute B determined in the plan design and the attribute H determined in the machine design are compared. If B> H is not satisfied, that is, if there is a contradiction in the constraint, the constraint C1 in the plan design is redone. there is a possibility. Therefore, the cycle detecting means 27 stores the path of C1, C3, C5, and C10 on the main storage as a path in which a cycle may occur. Next, the constraint priority determining means 28 determines the priority for each constraint belonging to the path related to the critical path, that is, the path of (C1 →) C3 → C4 → C6 in FIG. The importance data to be raised is set (405). Also, since the possibility of occurrence of a cycle is detected in the cycle detection (404), the importance of the constraint on the cycle is raised to the operator via the planning control unit 21 and the data output unit 23. Inquire whether or not. That is, an inquiry is made as to whether the importance of the constraint on the path of C1, C3, C5, and C10 should be increased (whether the occurrence of a cycle should be checked with the highest priority). The constraint priority determining means 28 raises the priority of the corresponding constraint according to the value of the operator's answer input via the data input means 22. FIG. 9 shows constraint importance data when the operator answers "do not increase importance" to the above-mentioned question, and FIG. 10 shows constraint importance data when the operator answers "increase importance". Shown respectively.
[0027]
By registering the importance data for each constraint set as described above in the design constraint storage means 41 for each of the design processes 503, 505, 507, and 509, the planning step ends (406).
[0028]
Next, the stage of executing the task of each design process will be described using the machine design process 504 as an example. The importance of the constraint will be described below in a case where the one shown in FIG. 9 is used. The machine designer starts the design based on the attribute B, which is the attribute information determined in the higher-level design process, that is, the plan design (410). First, the process support control unit 31 in the machine design workstation 504 activates the executable constraint extraction unit 35, determines which constraint in the design constraint storage unit 41 is executable from the set attributes, and starts the executable constraint extraction unit 35. The restriction is extracted (420 to 430). Executable constraints C8 and C3 are detected, so the constraint priority order judging means 36 is activated to compare the constraint importance data (440). Since the importance of the constraint C3 is 0.6 and the importance of the constraint C8 is 0.3, it is determined that the constraint C3 having the higher importance is to be activated (450). Next, the process support control unit 31 sends a command to the constraint propagation / work instruction unit 37 to activate the constraint C3 and calculate the value of the output attribute. The constraint propagation / work instructing unit 37 activates the local constraint propagation means 38 because the content of the constraint C3 is an arithmetic expression, and automatically outputs output attributes according to the operations D = f3 (B) and E = f4 (B). (480). The inconsistency detecting means 40 is activated based on the set attribute values, and checks whether the attribute values are inconsistent (510). Along with the setting of the attributes D and E, a constraint in which all related attributes have already been set does not appear newly, and contradiction is not detected. At this point, there are undetermined attributes F, H, and K (520), and the same processing is repeated. First, constraints C4, C5, and C8 are listed as startable constraints, and among them, C4 having the highest importance is selected and executed as an execution constraint. As a result, the value of the attribute F, which is an input attribute of the electric design, is calculated, and the process support control unit 31 transfers the value of the attribute F to the process support control unit 31 of the electric design workstation 506. Next, C5 is activated to determine the value of the attribute H. Since the content of the constraint C5 is an activity, the constraint propagation / work instructing unit 37 activates the work instructing unit 39 to generate the output attribute H. The activities ACT1 and ACT2 to be performed are displayed as work instruction items to the machine designer via the data output means 33 (490). The designer performs a design operation according to the instruction and sets the value of the output attribute H (500). The inconsistency detecting means 40 is activated based on the set attribute values, and checks whether the attribute values are inconsistent (510). Since the attributes H and B have been set, the constraint C10 can be activated, and it is verified whether B> H. When inconsistency is detected, the process support control unit displays a warning to the operator via the data output unit 33 (530). Finally, when the constraint C8 is activated and the value of the attribute K is determined, all the attributes related to the machine design process are determined, and the machine design process ends.
[0029]
Similarly, the electrical design process is performed in parallel with the mechanical design process, and finally the product prototype process is completed, thereby completing the entire design process.
[0030]
According to the present embodiment, the constraint propagation order of the mechanical design process, that is, the design procedure is guided so that the value of the input attribute F of the electrical design which is the critical path is determined with the highest priority. It can be seen that the specification waiting time in the electrical design C6, which is generated by starting C8 and the like first, can be minimized, and the total design period can be shortened.
[0031]
Further, since the planning function can be started at an arbitrary timing, a detailed plan (setting of the degree of constraint importance) can be successively reestablished in accordance with the progress of each design process.
[0032]
Furthermore, since the work procedure for performing each design work can be displayed and instructed as an activity, design efficiency can be guaranteed even when the skill level of the designer is low. For example, it can be realized by providing a linkable reference information of each activity, and providing a mechanism for searching from a design process workstation.
[0033]
【The invention's effect】
As described above, in a product design project in which the design is divided and shared among multiple design processes, the project management model is quickly generated based on the model of the design object in each design process. Can be done. In addition, using this model, it is possible to guide a design procedure that can reduce the design regression (cycle) and the waiting time for specification communication between design departments for each design process, that is, improve the cooperation efficiency. I can do it.
[0034]
In addition, since a detailed plan can be made sequentially according to the progress of the design project, it is possible to perform project management that can flexibly respond to changes in the project status.
[0035]
Furthermore, even when the skill level of the designer is low, a steady work instruction can be given, and the design period can be shortened.
[Brief description of the drawings]
FIG. 1 is a diagram showing a software configuration of a design support apparatus according to the present invention.
FIG. 2 is a diagram showing a hardware configuration of a design support apparatus according to the present invention.
FIG. 3 is a diagram showing a network configuration of a design support apparatus according to the present invention.
FIG. 4 is a flowchart showing a processing procedure of the present invention.
FIG. 5 is a diagram illustrating a network configuration according to an embodiment of the present invention.
FIG. 6 is a diagram showing constraint knowledge data for each design process in the embodiment of the present invention.
FIG. 7 is a diagram showing a constraint graph according to the embodiment of the present invention.
FIG. 8 is a diagram illustrating a critical path calculation result according to the embodiment of the present invention.
FIG. 9 is a diagram illustrating a calculation result (1) of the constraint importance in the embodiment of the present invention.
FIG. 10 is a diagram illustrating a calculation result (2) of the constraint importance in the embodiment of the present invention.
[Explanation of symbols]
11 Multibus,
12. Bus control device,
13. Central processing unit,
14: Main storage device,
15 ... Keyboard,
16 display devices,
17: Disk control device,
18 ... disk,
21: Planning control unit,
22 ... data input means,
23 ... data output means,
24 ... Project Management Department,
25 ... Constraint graph expansion means
26 ... Critical path calculation means,
27: cycle detection means,
28 ... constraint priority determining means,
31 ... Process support control unit,
32 ... data input means,
33 ... data output means,
34 ... design procedure guiding section,
35 ... executable constraint extraction means,
36 ... constraint priority order determination means
37 ... constraint propagation / work instruction section
38 ... local constraint propagation means
39 ... work instruction means,
40 ... contradiction detecting means
41 ... design constraint storage means
42 ... Required man-hour storage means.

Claims (4)

A design support method for supporting planning of an entire product design project at a planning stage of a product design project,
The product design project includes a plurality of design processes,
The design support method is realized using a computer including a data input unit, a data output unit, a main storage device, a central processing unit, and a storage unit,
The task of each design process, represented by the input attribute value and the output attribute value and the input attribute value and the relationship of the output attribute value, the input attribute value and the output attribute value and the relationship between the time required for the task Are stored in the storage means as constraint knowledge data,
The central processing unit,
When receiving input of an instruction for planning a product design project from the data input means,
Reading constraint knowledge data of a design process related to a product design project from the storage means ,
The input attribute value and the output attribute value of the read constraint knowledge data as a key to determine the precedence relationship of the tasks included in the design process,
Calculating a critical path in the process from the preceding relationship based on the required time of each task stored in the constraint knowledge data and the preceding relationship,
When the attribute value stored in the constraint knowledge data is provisionally determined, the attribute value provisionally determined in the upper-level design process is significantly larger than the precise attribute value obtained in the downstream design process. Differently, we checked the possibility of a cycle that would require changing all specifications determined based on the provisional value ,
Setting the importance to each task for each of the design processes based on the result of checking the possibility of a cycle due to the difference between the attribute result and the attribute value of the critical path,
The set importance is stored in the main storage device ,
Examples of the relationship between the input attribute value stored in the storage means and the output attribute value as a constraint knowledge data, work for obtaining the output attribute value from an arithmetic expression or input attribute values for obtaining an output attribute values from the input attribute value A design support method characterized in that a procedure is specified. A design support method for performing a task included in a plurality of design processes of a product design project drafted by the design support method according to claim 1, wherein the product design project management is performed to support design.
When the determined attribute value is input from the data input unit as input information,
The central processing unit,
Referring to the constraint knowledge data which have been stored in the storage unit to extract the tasks that can run the attribute value as the input attribute value,
If there are a plurality of extracted tasks, compare the importance stored in the main storage device, start the task according to the importance,
For the activated task, reference is made to the constraint knowledge data stored in the storage means, and if the relationship between the input attribute value and the output attribute value is an arithmetic expression, the output attribute value is calculated according to the arithmetic expression. When the relationship between the value and the output attribute value is a work procedure, the work procedure is controlled to be output to the data output means as a work instruction item,
An output attribute value of a result according to the work instruction output to the data output unit is received from the data input unit and set in the storage unit,
Design support method for designing assistance in the execution stage of the product design projects by is intended to make checks consistent with activation of tasks based on the calculated output attribute value or the output attribute value received. A design support device for supporting planning of the entire product design project in a planning stage of a product design project,
The product design project includes a plurality of design processes,
The design support device includes:
The task of each design process, represented by the input attribute value and the output attribute value and the input attribute value and the relationship of the output attribute value, the input attribute value and the output attribute value and the relationship between the time required for the task And storage means for storing as constraint knowledge data
The constraint knowledge data of the design process related to the product design project is read from the storage unit, and a constraint graph which is a precedence relation of a task included in the design process, using an input / output attribute value of the read constraint knowledge data as a key Constraint graph expansion means for obtaining
Critical path calculation means for calculating a critical path in the process from the preceding relationship based on the required time of each task stored in the constraint knowledge data and the preceding relationship,
If the attribute value stored in the constraint knowledge data is provisionally determined, the attribute value provisionally determined in the higher-level design process is significantly different from the precise attribute value obtained in the downstream design process, A cycle detecting means for checking the possibility of a cycle occurrence in which all the specifications determined based on the provisionally determined value have to be changed accordingly ,
Constraint priority determining means for setting a priority for each task for each of the design processes based on a result of checking the possibility of a cycle due to a difference between the result of identifying the critical path and the attribute value , and storing the priority in the storage means And having
The relationship between the input attribute value and the output attribute value stored in the constraint knowledge data, work procedures for the arithmetic expression or input attribute values for obtaining an output attribute values from the input attribute value obtaining output attribute value is defined A design support device characterized by: 4. The design support apparatus according to claim 3, wherein the configuration operates at a stage of executing a task included in a plurality of design processes of a product design project,
When the determined attribute value is the input means as the input information, and executable constraint extracting means for extracting a bootable tasks that attribute value as an input attribute values with reference to the constraint knowledge data,
When there are a plurality of extracted tasks, the stored priorities are compared, and a constraint priority order determining unit that starts the tasks according to the priorities;
When the relationship between the input and output attribute values of the activated task with reference to the constraint knowledge data is an arithmetic expression, a local constraint propagation unit that calculates an output attribute value according to the arithmetic expression,
A work instructing means for outputting the work procedure as a work instruction item when the relationship between the input and output attribute values is a work procedure, and receiving setting of an output attribute value as a result according to the instruction;
A design inconsistency detecting means for checking for inconsistencies based on the calculated output attribute values or the set output attribute values .

Images