JP5439296B2 - Change impact prediction method and change impact prediction apparatus - Google Patents

Description

  The present invention relates to data change management, and more particularly, to a technique for predicting in advance a range of other data that changes due to data change and its impact.

  In recent years, as the environment surrounding the manufacturing industry has become more severe, not only quickly developing and selling products that sell well, but also preparing for the accountability of manufacturers that comply with various regulations such as the environment, and post-sales services It is important to secure profits in the entire business including To that end, by managing product information at all stages from product planning and development to manufacturing, sales, maintenance, and disposal, it is possible to maximize profits and realize safety and security throughout the product life cycle. It becomes important. In this information management, it is important to make various data (functional specifications, part specifications, designs, books, procurement, manufacturing processes, test specifications, etc.) that make up the product data and manage the relevance of each data. is there. For example, when the design content of one part is changed, the influence range (specification of other parts that affect the specification, books, processes, etc.) is searched based on the information related to relevance. It is extremely important in project management that the project manager quickly grasps the exact impact (manual increase / decrease, cost increase / decrease, etc.). This is because there is a possibility that the decision on whether or not to carry out the design change is seriously affected depending on the impact. As described above, for the project manager, the change impact prediction for predicting the impact due to the change of the information related to the product is important not only in the design but at all stages.

  As a conventional technique related to change impact prediction in the project management, there is a method of calculating a change impact probability of each data that affects data when the data is changed using data connection accuracy (Patent Document 1). . Further, in the software component change support method, the first change type and the parameter are applied to determine whether to apply to the ripple change operation, and the second change type and the parameter that is the ripple change operation are derived. (Patent Document 2).

JP 2008-90784 A JP 2000-39998 A

  However, the above technique has the following problems.

  In patent document 1, in order to express the influence at the time of design change stochastically, the influence probability to all relevant information is calculated while gradually reducing the influence according to the connection order. In many cases, the presence or absence of an influence becomes deterministic depending on the content of the change, and it may be difficult to identify information that actually affects the change. Therefore, it is a problem to narrow down the influence range accurately by considering the original change contents.

  In Patent Document 2, a spillover change operation cannot be derived unless a combination of a first change type to be changed and a change parameter (change object and change content) exists in the past. For example, even if there is a track record of a combination with a certain change parameter for a certain change type, such a change request cannot be handled without a combination with another change parameter. In particular, when making a change impact prediction for product information that does not necessarily have high repeatability and reusability, such as products with complex configurations and made-to-order products, the change parameters are different even if the change type is the same. Therefore, it is a problem to accurately predict the influence of the change even when there is no past change record.

  The present invention has been made in order to solve the above-mentioned problems, and its purpose is that the change type is often the same even if the change type is the same. The purpose is to generate a pattern that can narrow down with high accuracy.

  In order to solve the above problem, the change impact prediction method of the present invention is a change impact prediction method by a change impact prediction apparatus having a change history database that is past history information about changes, and the change history database identifies changes. Change ID, change content, and other change ID affected by the change as a variable of change information, and the change impact prediction device includes one or more variables of change information constituting the change record database. A change type determining step for determining a change type by accepting a selection of a variable; a classification step for classifying the change history of the change record database according to the change type according to the change type; a change ID on the change generation side; A combination of the change type on the occurrence side and the change type on the influence side using the change ID on the influence side affected by the change And a change pattern generation step of generating a modified pattern.

  According to the present invention, since the change parameters are often different even if the change type is the same, it is possible to generate a pattern that can narrow down the influence range with high accuracy by considering the original change content.

It is a block diagram of the change influence prediction apparatus concerning one Embodiment of this invention. It is an example of change performance DB11. It is an example of a component classification and a site | part. It is an example of a change content. It is a flowchart explaining the content of the change classification determination process. It is an example of change statistics DB. It is an example of change pattern DB. It is an example of GUI for inputting change information. It is an example of component DB. It is a flowchart explaining the content of the process in an influence estimation part. It is an example of the node table format of relevance network DB. It is an example of the link table format of relevance network DB. It is an example of the book table format of relevance network DB. It is an example of the parts table format of relevance network DB. It is an example of the schematic diagram of relevance network DB. It is an example of a change influence list. It is an example of change influence prediction result totaling / display. It is an example of the change influence prediction result using a three-dimensional CAD system.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a configuration diagram of a change impact prediction apparatus according to an embodiment of the present invention. The change impact prediction apparatus includes a change result processing unit 10 and a change impact prediction unit 14. The change record processing unit 10 determines a change type using information in the past change record DB 11 and creates a change statistics DB 12 and a change pattern DB 13 including change impacts (addition man-hours, additional costs) for each change type. . When the change impact prediction unit 14 inputs the contents when a new change case occurs, information about the relevance network DB 15 in which the relevance between the change statistics DB 12 and the change pattern DB 13 and the data is defined and information about the parts are obtained. Using the stored component DB 17, the influence range (affected data) and the impact with respect to the input change are calculated as predicted values and output to the display device 16 such as a CRT, LCD, PDP or the like.

  In the following, in order to explain this embodiment more specifically, the design information of the plant will be described as an example of the target product. However, the present invention is not necessarily limited to this, and in addition to construction structures such as buildings and bridges, It may be applied to information on various products such as hardware such as products and electrical products, and software.

  Next, the function of each part which comprises the said change influence prediction apparatus is demonstrated.

  The change result processing unit 10 includes a change type determination unit 100 and a change result data processing unit 101, determines a change type using information in the change result DB 11, and changes impact (addition man-hour, additional cost) for each change type. ) Including the change statistics DB 12 and the change pattern DB 13.

  The change record DB 11 stores change unit history information regarding past similar products including the product.

  Here, FIG. 2 shows an example of the change record DB 11. As shown in the figure, when a change item such as a design change occurs in the same type of product in the past, the contents are saved for each item. The configuration includes parameters such as a change ID for identifying an item, a change name, a product ID, a component ID, a part, a change content, a phase, an impact (additional man-hours, additional cost), and an influence side change ID. These are change information variables. Here, the parameters will be described. The change ID is a number for identifying the change and is created for each part to be changed. The product ID is an ID for identifying a product (for example, plant A) including the change (part). As shown in FIG. 3, the part type and part are assigned a type number for each part (including books), and if there are subparts belonging to the part, the part (subpart type) is assigned. Assigned. The component ID is a number for identifying each component, and a number is assigned for each component type. As illustrated in FIG. 4, the change content is assigned a number for each change content. The phase is a progress stage of the target part at the time of occurrence of the change, and may be set according to the relationship with the impact. For example, if it is 1 after the design details are confirmed and before parts procurement (ordering) and 2 if it is after parts procurement, the correlation with the cost impact will be particularly significant when statistical processing is performed later. Can be increased. Regarding impact, there are two types of additional man-hours and additional costs, which are respectively the man-hours (man-hours) and costs (k ¥) additionally generated for the change ID. As for the additional man-hour and the additional cost, a negative value is obtained when the change is less than the original schedule due to the change, a zero value is obtained when there is no change, and a positive value is obtained otherwise. The impact may be an absolute value of man-hours and costs, but since differences such as part size are included as influencing factors, it is normalized by the quantity (for example, weight) possessed for each part ID. It is desirable to be a thing. In the following, impact is treated as normalized by physical quantity (ie per unit physical quantity).

  The influence side change ID is another change ID generated by the change. For example, the influence side change IDs 202 and 203 of the change ID 201 mean that two changes of the change IDs 202 and 203 have occurred along with the change of the change ID 201. In other words, the change ID 201 is the side that caused the change (occurrence side), and the change IDs 202 and 203 are the side that was affected by the change (influence side). Further, the change ID 207 is an influence side with respect to the change ID 206 and a generation side with respect to the change ID 208.

  Next, the function of each part which comprises the change performance process part 10 is demonstrated.

  When creating the change statistics DB 12 based on the change record DB 11, the change type determination unit 100 determines a change type that is a creation unit by selecting one or more parameters from the parameters constituting the change record DB 11. By the way. In change type determination, that is, parameter selection, the selection may be made based on the experience of the change manager, or may be made by quantitative evaluation.

  Here, an example of parameter selection by quantitative evaluation will be described with reference to the flowchart of FIG.

  First, the change record DB 11 is read (S50).

  Next, the parameters constituting the change record DB 11 (in the example of FIG. 2, change ID to influence side change ID) are extracted, and an arbitrary combination of parameters (parameter set) is created (S51). However, as described above, since the parameter set is to be a creation unit for creating the change statistics DB 12, a parameter set that is not related to the parameter set is excluded from candidates in advance, or a parameter that is always included is specified. Restrictions may be added. For example, in FIG. 2, there are 31 parameter sets including one or more of five parameters of product ID, part type, part, change content, and phase. Of the five parameters, for example, there are fifteen types of constraint conditions that always include the component type.

  Next, according to a certain parameter set i, each change in the change record DB 11 is totaled, and statistical processing is performed regarding impact (additional man-hours, additional cost) (S52). For example, when the parameter set i is composed of two parameters, that is, the component type and the change content, the impact of the change ID is statistically processed in units of a combination j (referred to as a change type) of the two parameters. In the example of FIG. 2, in the change IDs 202, 203, and 205, the component type is 2 (piping) and the change content is the same as 2 (route, dimension change). , Additional cost) and statistical processing. As a result of the statistical processing, the average value, median value, variance (standard deviation), frequency, and the like are output. This statistical processing is performed for all combinations j (part type 1 and change content 1, part type 1 and change content 2,...) In the change type.

  Next, using the statistical value for each parameter set i and change type j calculated in S52, an evaluation value Ei for the parameter set i is calculated (S53). The evaluation formula for calculating the evaluation value Ei is defined by the one relating to the variation in impact for each change type, for example, as in the following formula.

Here, M _ij and C _ij are the additional man-hours and additional costs for the parameter set i and the change type j, respectively, and S (M _ij ) and S (C _ij ) are the additions for the parameter set i and the change type j, respectively. Standard deviation of man-hours and additional costs. α is a coefficient for converting man-hours into costs (dimension is ¥ / hour). For example, the charge or self-sufficiency of the project may be applied. m is the number of change types j. However, if there is even one change type whose data frequency is less than the predetermined number, the evaluation value Ei is not calculated because the statistical reliability is low (the maximum value that can be taken programmatically is assigned). .

  Next, it is determined whether or not evaluation values have been calculated for all the parameter sets created in S51. If Yes, the process proceeds to S55, and if No, the process returns to S52 and continues (S54).

  Finally, the smallest parameter set i among the evaluation values calculated for each parameter set is determined as a creation unit for creating the change statistics DB 12. If the parameter set is not uniquely determined in this step (for example, when the evaluation values of a plurality of parameter sets are minimum values), the evaluation value for each parameter set may be presented to the user.

  In this way, the change type determination unit 100 can determine an appropriate parameter set.

  The change record data processing unit 101 processes data in the change record DB 11 in accordance with the parameter set determined by the change type determination unit 100. Regarding data processing, there are processes for generating the change statistics DB 12 and the change pattern DB 13. Next, each process will be described.

  The change statistics DB generation process performs the same process as S52 in the parameter selection flowchart (FIG. 2) based on the quantitative evaluation, and outputs the result to the change statistics DB 12. Here, FIG. 6 shows an example of the change statistics DB 12 generated based on the change record DB 11 of FIG. 2 when the determined parameter set is composed of four parameters of part type, part, change contents, and phase. . The change type is an ID for identifying with the four parameters. The change type name is a combination of the names of the parameters. The impact is an average value or a median value for each change type. However, as described above, it may be an average value or a median value regarding what is once normalized by an amount (for example, weight).

  The change pattern DB generation process extracts a combination of a change type corresponding to each of the change ID and the influence side change ID in the change record DB 2, thereby changing the change pattern composed of the combination of the change type on the generation side and the change type on the influence side. And outputs the result to the change pattern DB 13. An example of the generated change pattern DB 13 is shown in FIG. Here, the change pattern generation processing will be described with reference to the examples of FIGS. The change ID 201 corresponds to the change type 1 from the four parameters (part type, part, change content, phase). Therefore, the change type on the generation side is 1. The change type is 3 for both 202 and 203 which are the influence side change IDs of the change ID 201. Therefore, the change type on the influence side at this time is 3. By applying the above process to all the data included in the change record DB 11 and classifying the data, a change pattern including a combination of the change type on the generation side and the change type on the influence side can be generated. Then, the occurrence rate of the change type on the influence side with respect to the change type on the generation side is obtained from each frequency included in the change statistics DB 12. However, a phase is not considered in the frequency at which the occurrence rate is obtained.

  As described above, a change impact prediction method using a change impact prediction apparatus having a change history database that is past history information regarding a change, wherein the change history database includes a change ID for identifying a change, a change content, Another change ID affected by the change is included as a variable of change information, and the change impact prediction device receives the selection of one or more variables from the change information variables constituting the change record database, thereby changing the change type. A change type determination step to be determined, a classification step of classifying the change history of the change record database according to the change type by the change type, a change ID on the change generation side, and an influence side on which the change has affected A change that uses a change ID to generate a combination of the change type on the occurrence side and the change type on the influence side as a change pattern By performing the change impact prediction method with a turn generation step, the change parameters are often different even if the change type is the same, so the impact range can be narrowed down accurately by considering the original change contents A pattern that can be generated can be generated.

  In addition, as a change impact prediction device, there is a change record database that is past history information regarding a change, and the change record database includes a change ID for identifying a change, a change content, and other changes affected by the change. A change type determining means for determining a change type by accepting selection of one or more variables from a variable of change information constituting the change record database, including a change ID as a variable of change information; and the change according to the change type Using the classification means for classifying the change history of the results database by the change type, the change ID of the change occurrence side, and the change ID of the influence side affected by the change, the change type and influence side of the occurrence side The change type is the same by including a change pattern generation means for generating a combination of change types as a change pattern. Because there are many that change parameters are different, it is possible to generate a pattern which can be narrowed accurately influence range by considering the original changes.

  In addition, the change record database further includes, as variables, additional man-hours required for the change, additional cost, and phase information regarding the progress stage of the target component at the time when the change is made, and in the change type determination step, the phase information Is further input as a classification condition for change impact calculation, and in the classification step, statistical values of additional man-hours, additional costs, and occurrence frequency required for the change are calculated and changed according to the change type and the phase information. You may perform the change impact prediction method stored in a statistical database. As a result, in the change management of various information (data) related to products, especially products that do not have high repeatability and reusability of information, the correlation of the change impacts is ensured against the past change results including change impacts. By categorizing so as to perform, it is possible to improve the applicability of the change record data to the change impact prediction.

  Further, as the change impact prediction device, the change record database further includes, as variables, additional man-hours required for the change, additional cost, and phase information regarding the progress stage of the target part at the time when the change is made, and the change type determination In the means, the phase information is further received as a change impact calculation classification condition, and the classification means obtains the statistical values of additional man-hours, additional costs, and occurrence frequency required for the change in accordance with the change type and the phase information. By calculating the change impact and storing it in the change statistics database, in the change management of various information (data) related to products, especially products that do not have high repeatability and reusability of information, it is possible to obtain past change results including change impacts. On the other hand, categorize to ensure the correlation of the change impact. It is possible to improve the applicability to the change impact prediction changing actual data by.

  In addition, the change type determination step creates a parameter set composed of one or more arbitrary number of parameters among a plurality of parameters, and statistical variation regarding the additional man-hour and the additional cost of the change record database for each parameter set. The change type may be determined by evaluating and selecting the parameter set having the least variation. Thereby, it is possible to evaluate the validity of the change by calculating a quantitative impact such as man-hours and costs caused by the change.

  In addition, as the change impact prediction device, the change type determination unit creates a parameter set including one or more arbitrary number of parameters among a plurality of parameters, and the additional man-hour and the additional cost of the change record database for each parameter set By evaluating the statistical variation and determining the type of change by selecting the parameter set with the least variation, the validity of the change is calculated by calculating the quantitative impact such as man-hours and costs caused by the change. Can be evaluated.

  Next, the change impact prediction unit 14 will be described. The change impact prediction unit 14 includes a change information input unit 140, an impact prediction unit 141, and a prediction result output unit 142, and determines the change pattern by searching the change pattern DB 13 based on the input change information. In addition, the influence range and impact from the input change are predicted from the relevance network DB 15 including the relevance between the change statistics DB 12 and data, and the prediction result is output to the display device 16.

  Next, the function of each part which comprises the change influence estimation part 14 is demonstrated.

  The change information input unit 140 is where the user inputs information related to the change contents. As an example, input by a GUI (Graphical User Interface) shown in FIG. The user inputs the contents of change by inputting the contents of the displayed reference numerals 80 to 85 which are displayed parameters through a GUI operating on a system such as a PC. For the component type 80 (component ID) and the part 81, parameters of the component DB 17 illustrated in FIG. 9 may be searched and input by keyword search or the like. In the case where the options are predetermined as in the change contents 82 and the phase 83, all the options are displayed in the pull-down menu by inputting the parameter input areas 86 and 87 with an input device such as a mouse. The user may input by selecting from among them. For the additional man-hour 84 and the additional cost 85, numerical values are input in the input area. After all the input is completed, the input of the change information is completed by pressing the “OK” button with an input device such as a mouse. When the “Cancel” button is pressed, information already input in this GUI is deleted.

  The impact prediction unit 141 uses the change statistics DB 12, the change pattern DB 13, and the relevance network DB 15 to determine the impact range and impact of the change based on the information regarding the change content input by the change information input unit 140. I am predicting. The specific processing contents in the influence prediction unit 141 will be described with reference to the flowchart of FIG.

  First, in order to read data from the change statistics DB 12, the change pattern DB 13, and the relevance network DB 15, the connection is established to these DBs, and the connection is established through authentication of the user name and password (S60).

  Next, the generation side change type is specified from the change information input by the change information input unit 140, and the impact of the generation side change is predicted (S61). In the following description, as an example of the change type, a case will be described in which the parameter is composed of four parameters: a part type, a part, a change content, and a phase. The change statistics DB 12 illustrated in FIG. 6 is also based on this change type parameter. When the change information input by the change information input unit 140 is as illustrated in FIG. 8, the change type can be identified as 1 by comparing the change type number in FIG. 6 with each parameter. If the change type cannot be specified, the change type is set to 0 (meaning that cannot be specified). Next, the impact of the change on the generation side is predicted. From FIG. 6, the additional man-hours and additional costs per unit quantity on the generation side are 4 [person-hour / kg] and 50 [k ¥ / kg], respectively. Here, it is found that the nozzle 1 of the tank is the change target by searching the component DB 17 using the part type, the part ID, and the part (1, 1, 1 respectively) of the change target part on the generation side as a key. . Further, since the weight is 20 [kg], the estimated value of the additional man-hour and the additional cost is calculated by applying this weight to the additional man-hour and the additional cost per unit quantity. Here, for the sake of simplification, it is assumed that the prediction value is simply a prediction model in which the prediction value is simply proportional to the weight. It becomes. In addition, about a prediction model, it should not be fixed to the said model, and in order to improve prediction accuracy, it may replace with another model for every component classification. Further, by referring to the relevance network DB 15 described later, the node ID corresponding to the change target part on the generation side is specified, and is registered in the change influence list illustrated in FIG. 16 together with the impact. Furthermore, a book related to the part to be changed is extracted and registered in the change influence list together with the changed part (if the book cannot be extracted, it is not registered). The impact prediction value of the book is set to 0 because it is included in the predicted value of the part.

  Next, the change pattern DB 13 is searched, and it is determined whether there is a change pattern in which the specified occurrence side change type matches the change type (occurrence side) in the change pattern DB 13 (S62). In the above example, the identified change type on the occurrence side is 1, and it is found that the change pattern A matches with the change type (occurrence side) in the change pattern DB of FIG. Therefore, the determination result in this case is Yes, and the process proceeds to S63. If there is no matching change pattern, the determination result is No and the process proceeds to S68. In this step, since only the process on the generation side is performed and the process on the influence side is not performed, it is assumed that the change influence order indicating the number of stages of influence due to the change is zero.

  When the determination result in S62 is Yes, the change influence order is incremented by 1 (when the change influence order so far is 0, it is 1). Also, the change pattern (influencing side) corresponding to the change pattern is acquired by referring to the change pattern DB 13 (S63). In the above example, it is the change pattern A, and it can be seen that the change type (influence side) in this case is 3.

  Next, it is determined whether the influence side node, the change type, and the impact can be predicted (S64).

  Here, since the relevance network DB 15 illustrated in FIGS. 11 to 15 is referred to, these will be described first. FIG. 11 shows an example of a table format for defining the nodes constituting the network. Parts, books, etc. are distinguished as node types. The detailed attribute of the node is linked to another table (FIGS. 13 and 14) designated by Object_Type (node type) and Object_ID. FIG. 12 shows an example of a table format for defining the links constituting the network. The link direction is defined by defining a parent node (upstream node) and a child node (downstream node). FIG. 13 shows an example of a table format for storing information related to books. When the node type is a book, it is referred to and attributes shown in the figure are defined. FIG. 14 shows an example of a table format for storing information related to parts. The table is referred to when the node type matches the part, and the detailed attribute of the part is the part DB 17 according to Object_Type (part type) and Object_ID. Linked to (Fig. 9). Here, a schematic diagram is shown in FIG. 15 as an example of the relevance network DB 15. ○ and □ indicate nodes, and the numbers are node IDs. The arrows indicate the link and the direction of influence. If one node is changed, it means that there is a possibility of affecting the node indicated by the arrow coming out of the node. In FIG. 15, having two links from node 1 to node 6 means that there are two paths connecting both nodes because there are a plurality of parts.

  Returning to S64, the description will be continued. Since it was found that the change type (influencing side) is 3 in S63, the part type is 2 (piping) and the change content is 2 (route, dimension change) by referring to the change statistics DB in FIG. I understand. Here, a description will be given using the relevance network of FIG. Assume that the generation target component is a node 1. At this time, the candidates for the affected parts are nodes 2, 3, 4 and 6 (the book is excluded because it is not a part). It is determined whether or not the prediction is possible depending on whether the candidate node part type matches the pipe that is the obtained influence-side part type and matches the generation side change part. If there is no match (No), the process proceeds to S68. For example, if the nodes 2 and 6 satisfy the condition, the determination becomes Yes and the process proceeds to S65. In this way, by checking whether the part type and part on the change generation side and the influence side are consistent with the change pattern, the nodes that may be affected are appropriately narrowed down.

  When it is determined Yes in S64, each node (part) that matches the condition and the related book are registered in the change impact list, the impact is predicted for each node, and the prediction result is changed. It is registered in the influence list (S65). In the above example, since the nodes 2 and 6 match the conditions, the prediction is made in the same manner as the impact prediction of S61 from the physical quantity (obtained from the parts DB 17) of each node (piping) and the impact per unit physical quantity (obtained from the change statistics DB 12). Calculate. The impact of the book is set to 0 because it is included in the part.

  Next, with reference to the change pattern DB 13, it is determined whether or not there is one having the change type on the influence side (3 in the above example) as the change type on the generation side (S 66). In the example of the change pattern DB in FIG. 7, since the pattern F is applicable, it is determined as Yes and the process proceeds to S67. If it does not exist, it is determined No, and the process proceeds to S68.

  When it is determined Yes in S66, it is determined whether or not the change influence order exceeds a predetermined value (S67). This process is for the purpose of preventing the influence of the change from spreading unnecessarily, and the predetermined value may be specified from the experience of the change or may be generated by analyzing the results of the change record DB 11. A change influence order by change may be obtained, and a numerical value based on the past change record such as the maximum value or the 95th percentile value may be adopted. When it determines with Yes, it progresses to the process of S68. When it determines with No, it returns to the process of S63 and repeats the process of S63-S66, and the influence prediction by change is performed.

  When it is determined No in S67, it is determined whether all the influence side change types have been processed (S68). This is to cope with the case where the influence of the change extends to a plurality of nodes. When it determines with Yes, it progresses to the process of S69, and when it determines with No, it returns to the process of S63 and repeats the process of S63-S66, and the influence influence by a change is predicted.

  When it is determined Yes in S68, the change impact prediction result is totaled with reference to the change impact list (S69). Specifically, the impact predicted values (additional man-hours and additional costs) included in the change impact list are summed up. Alternatively, as illustrated in FIG. 17, image data of a change influence tree may be created for each node. In this tree image, the predicted impact value 90 for each node and the occurrence rate 91 for each link (obtained from the change pattern DB 13) may be written together. Further, the display form of nodes and links may be changed according to the predicted impact value and the occurrence rate. Examples of changing the node display mode include changing the line thickness and line type (solid line / broken line, etc.) of the figure surrounding the node number, changing the size of the figure surrounding the node number, or changing the node number. Coloring (hatching) the surrounding figure. As an example of changing the display form of the link, changing the thickness or line type of an arrow representing the link can be cited.

  In this way, the impact prediction unit 141 can predict the range and impact of the change input by the change information input unit 140. Further, when the user has a plurality of proposals to be changed, the impact prediction value for each change proposal can be obtained by inputting each of them with the change information input unit 140, and the impact prediction values are compared. It is also possible to make a qualitative assessment of the proposed change (the smaller the impact, the better the quality).

  The prediction result output unit 142 reads the information acquired by the influence prediction unit 141 and the output information and transmits a display signal to the display device 16 in the form of text or graphics. Accordingly, for example, a change impact list illustrated in FIG. 16 or a change impact tree illustrated in FIG. 17 can be displayed on the display device 16. Alternatively, instead of the tree of FIG. 17, as illustrated in FIG. 18, by using a three-dimensional CAD system, the generation-side changed part object 92 and the affected-side part object 93 are colored with different colors, A specific influence range related to the component can be presented to the user.

  In this way, in the change management of various information (data) related to products, the applicability of change record data to change impact prediction can be improved, and the impact range and impact due to the change can be predicted with high accuracy. it can.

  As described above, the change impact prediction device uses the change information input step for inputting the change information on the generation side and the change type of the change information determined in the change type determination step. A step of identifying a change type, and a search for and extracting a change pattern in which the change type of the occurrence side in the change pattern corresponds to the specified change type, and extracting the influence side change type in the extracted change pattern, An impact prediction step of calculating the impact of the change by searching the change statistics database based on the occurrence side change type and the influence side change type may be further provided. Thereby, it becomes possible to predict the influence range and impact due to the change with high accuracy.

  Further, as the change impact prediction device, the change information input means for inputting the change information on the generation side, and the change information input using the correspondence between the change information determined by the change type determination means and the change type A means for specifying a change type; and a change pattern in which the change type on the occurrence side in the change pattern corresponds to the specified change type is extracted, and an influence side change type in the extracted change pattern is extracted, and Based on the change type on the occurrence side and the change type on the influence side, an impact prediction means for calculating the impact of the change by searching the change statistics database is provided, thereby predicting the impact range and impact due to the change with high accuracy. Will be able to.

  As described above, the change impact prediction apparatus has a relevance network database that represents the relevance of individual parts or books, and in the impact prediction step, the parts that are related to the occurrence side change from the relevance network database. May be extracted as a candidate, and an influence range may be determined by extracting a part type represented by the influence side change type and a change content that matches the candidate. Thus, by narrowing down the predicted influence range in consideration of the relevance between data, it becomes possible to predict the influence range and impact due to the change with high accuracy.

  In addition, the change impact prediction apparatus has a relevance network database representing the relevance of individual parts or books, and the impact prediction means extracts parts related to the occurrence side change from the relevance network database as candidates. Then, by extracting the part type indicated by the influence side change type and the change content that matches the candidate, the influence range is determined, and the predicted influence range is narrowed down in consideration of the relationship between the data. By doing so, it becomes possible to predict the influence range and impact of the change with high accuracy.

  As described above, the change impact prediction device is input by repeating extraction of the influence side change type with the influence side change type extracted corresponding to the occurrence side change type as a new occurrence side change type. There may be a step of creating change influence tree information including the contents of the chained change from the changed information. As a result, a plurality of change proposals can be visually examined.

  Also, as a change impact prediction device, by repeatedly extracting an impact side change type as an impact side change type extracted corresponding to the occurrence side change type as a new occurrence side change type, from the input change information By providing the means for creating the change impact tree information including the contents of the chained changes, it is possible to visually examine a plurality of change plans.

  As described above, the change tree information is defined as a part or a book related to an occurrence side change and an influence side change as a node, and a relationship between the nodes is defined as a link. Display that displays in a tree composed of nodes and links using the contents of the chained changes from, and outputs to the display device by distinguishing the display form of the nodes or links according to the magnitude or occurrence rate of the impact of the changes A step may be further provided. This makes it possible to perform a qualitative evaluation by comparing the impact prediction values of multiple proposals, and can effectively assist in narrowing down proposals that should be formally adopted. Efficiency can be promoted. Further, as a change impact prediction device, the change tree information is defined as a component or a book related to an occurrence side change and an impact side change as a node, and a relationship between the nodes as a link. Displayed in a tree composed of nodes and links using the contents of the change, and provided with display control means for distinguishing the display form of the nodes or links according to the magnitude or occurrence rate of the change and outputting to the display device As a result, it becomes possible to perform a qualitative evaluation by comparing the impact prediction values of multiple change proposals, and can effectively support the narrowing down of proposals that should be officially adopted. Efficiency can be promoted.

  In the above-described example, it is described that the tree display in FIG. 17 is created by performing the processing in FIG. 10 in the impact prediction unit 141. However, the change impact list data in FIG. The data such as the change pattern database of FIG. 7 may be created in the database, and this may be displayed in the tree display of FIG. In other words, it has a database that defines parts or books related to originating side changes and impacting side changes as nodes and links between the nodes as links, and associates them with parts or books related to originating side changes and impacting side changes. It has a database that records the magnitude or occurrence rate of the impact of the change, is displayed in a tree composed of the nodes and the links, and the display form of the node or link according to the magnitude or the occurrence rate of the impact of the change It is good also as a display method of the change influence support apparatus provided with the display control step which distinguishes and outputs to a display apparatus. This makes it possible to perform a qualitative evaluation by comparing the impact prediction values of multiple proposals, and can effectively assist in narrowing down proposals that should be formally adopted. Efficiency can be promoted. The database may record the impact of the additional man-hour or the additional cost and the node, and the frequency or the occurrence rate and the link in correspondence.

  In addition to or separately from the display that distinguishes the display form described above, the expected impact value to the end of the tree is calculated using the impact and incidence of additional man-hours or additional costs from the upstream of the tree, and the tree By displaying in correspondence with the end node, it is possible to effectively assist in narrowing down the proposed change. Further, not the expected value of impact up to the end of the tree but the expected value of impact up to a node in the middle of the tree may be calculated and displayed. Moreover, it is good also as a display which distinguished the display form of the node and link mentioned above corresponding to an expected value.

  Further, the above tree display can be applied to the CAD system display of FIG. 18, and the change part object 92 on the generation side and the part object 93 on the influence side are colored with different colors, or the expected value of impact is calculated. The expected value may be displayed in correspondence with the affected part object. The user can be presented with a specific range of influence and degree of influence on the part.

  In the above-described example, the processing method of the change impact support apparatus is described. However, the same effect can be obtained by a change impact support apparatus having a processing unit that performs each process.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them, for example, with an integrated circuit. Further, each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files that realize each function can be stored in a memory, a recording device such as a hard disk or an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD. Therefore, each process and each configuration can realize each function as a processing unit or a program module.

  The present invention is not limited to the above-described embodiments, and includes various modifications. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

  Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

  As described above, the change impact prediction method and the change impact prediction apparatus according to the present invention provide change impact in change management of various information (data) related to products, particularly products that are not highly repeatable and reusable. In addition to improving the applicability of change history data to change impact predictions by categorizing the past change results to ensure the correlation of the change impacts, the impact of predictions is also considered in consideration of the relationship between the data. By narrowing down the range, it becomes possible to predict the affected range and impact due to the change with high accuracy. In addition, it becomes possible to perform a qualitative assessment by comparing the impact prediction values of multiple change proposals, which can effectively assist in narrowing down the change proposals that should be officially adopted. Has the effect of being able to promote efficiency in business, for example, various engineering systems related to product development and various computer systems such as PCs, servers, mobile information terminals connected to the management system via a network, The present invention can be applied to product data management software that operates on a computer system.

10 Change record processing unit 11 Change record DB
12 Change statistics DB
13 Change pattern DB
14 Change impact prediction unit 15 Relevance network DB
16 Display device 80 Part type 81 Part 82 Change contents 83 Phase 84 Additional man-hour 85 Additional cost 86, 87 Parameter input areas 90a, 90b, 90c Impact predicted values 91a, 91b for each node Incidence rate 92 for each link Occurrence side Change part object 93 influence side part object 100 change type determination unit 101 change result data processing unit 140 change information input unit 141 influence prediction unit 142 prediction result output unit

Claims (16)

A change impact prediction method by a change impact prediction apparatus having a change history database that is past history information regarding changes,
The change record database includes a change ID for identifying a change, a change content, and another change ID affected by the change as a variable of change information,
The change impact prediction device includes:
A change type determination step of determining a change type by accepting selection of one or more variables from variables of change information constituting the change record database;
A classification step of classifying the change history of the change record database according to the change type by the change type;
A change pattern generation step of generating a combination of the change type of the generation side and the change type of the influence side as a change pattern using the change ID of the change generation side and the change ID of the influence side affected by the change; A change impact prediction method characterized by comprising: The change record database further includes, as variables, additional man-hours required for the change, additional cost, and phase information regarding the progress stage of the target part at the time when the change is made,
In the change type determination step, the phase information is further accepted as a change impact calculation classification condition,
In the classification step, statistical values of additional man-hours, additional costs, and occurrence frequencies required for the change according to the change type and the phase information are calculated as change impacts and stored in a change statistics database. 1. The change impact prediction method according to 1. The change type determination step creates a parameter set composed of an arbitrary number of one or more parameters among a plurality of parameters, and evaluates the statistical variation with respect to the additional man-hour and the additional cost of the change record database for each parameter set. 3. The change impact prediction method according to claim 2, wherein the change type is determined by selecting the parameter set having the least variation. The change impact prediction device includes:
A change information input step for inputting change information on the originating side;
Identifying the change type of the input change information using the correspondence between the change information and the change type determined in the change type determining step;
The change pattern in which the change type in the change pattern corresponds to the specified change type is searched and extracted, the change type in the change side in the extracted change pattern is extracted, and the change type in the generation side and the influence side are extracted. The change impact prediction method according to claim 2, further comprising an impact prediction step of calculating an impact of the change by searching the change statistics database based on a change type. The change impact prediction device includes:
A relevance network database that represents the relevance of individual parts or books;
In the impact prediction step, by extracting a part related to the change on the occurrence side as a candidate from the relevance network database, and extracting a part type represented by the influence side change type and a change content that matches the candidate 5. The change impact prediction method according to claim 4, wherein an impact range is determined. The change impact prediction apparatus repeats extracting the impact side change type as the impact side change type extracted corresponding to the occurrence side change type as a new occurrence side change type, thereby obtaining the change side from the input change information. The change impact prediction method according to any one of claims 1 to 5, further comprising a step of creating change impact tree information including contents of chained changes. The change tree information is defined by defining parts or books related to the change on the occurrence side and the change on the influence side as nodes, and the relationship between the nodes as a link,
The change impact prediction device displays a tree of nodes and links using the contents of the chained change from the input change information, and changes the node or link according to the magnitude or occurrence rate of the change impact. The change impact prediction method according to claim 6, further comprising a display control step of distinguishing display forms and outputting to a display device. It has a change record database that is past history information related to the change, and the change record database uses a change ID for identifying a change, a change content, and another change ID affected by the change as a variable of the change information. Including
A change type determining means for determining a change type by accepting selection of one or more variables from variables of change information constituting the change record database;
Classification means for classifying the change history of the change record database according to the change type by the change type;
Change pattern generation means for generating, as a change pattern, a combination of the change type on the occurrence side and the change type on the influence side using the change ID on the change occurrence side and the change ID on the influence side affected by the change; A change impact prediction apparatus characterized by comprising: The change record database further includes, as variables, additional man-hours required for the change, additional cost, and phase information regarding the progress stage of the target part at the time when the change is made,
In the change type determination means, the phase information is further accepted as a change impact calculation classification condition,
The said classifying means calculates each statistical value of the additional man-hour, additional cost, and occurrence frequency required for the change according to the change type and the phase information as a change impact and stores it in a change statistics database. 8. The change impact prediction apparatus according to 8. The change type determining means creates a parameter set composed of an arbitrary number of one or more parameters among a plurality of parameters, and evaluates the statistical variation regarding the additional man-hour and the additional cost of the change record database for each parameter set. 10. The change impact prediction apparatus according to claim 9, wherein the change type is determined by selecting the parameter set having the least variation. Change information input means for inputting change information on the generation side;
Means for identifying the change type of the input change information using the correspondence between the change information and the change type determined by the change type determination means;
The change pattern in which the change side in the change pattern corresponds to the specified change type is searched and extracted, the change type in the change side in the extracted change pattern is extracted, and the change type in the generation side and the influence side are extracted. 11. The change impact prediction apparatus according to claim 9, further comprising impact prediction means for calculating an impact of the change by searching the change statistics database based on a change type. A relevance network database that represents the relevance of individual parts or books;
In the influence prediction means, by extracting a part related to the change on the occurrence side as a candidate from the relevance network database, and extracting a part type indicated by the influence side change type and a change content that matches the candidate The change influence prediction apparatus according to claim 11, wherein an influence range is determined. Includes the contents of the chained change from the input change information by repeating the extraction of the influence side change type with the influence side change type extracted corresponding to the occurrence side change type as a new occurrence side change type The change impact prediction apparatus according to claim 8, further comprising means for creating change impact tree information. The change tree information is defined by defining parts or books related to the change on the occurrence side and the change on the influence side as nodes, and the relationship between the nodes as a link, and using the contents of the chain change from the input change information, The display apparatus according to claim 1, further comprising: a display control unit configured to display a tree of links, and display a display form of the nodes or links in accordance with the magnitude or occurrence rate of the change impact. Item 14. The change impact prediction device according to Item 13. It has a database defined with the parts or books related to the change on the originating side and the change on the affecting side as nodes and the relationship between the nodes as links.
A method of displaying the change impact support device which have a size or database recording the incidence of impact of the change in association with generation side changed parts or books relating to the impact side changes,
The change impact prediction device includes:
Display as a tree consisting of the nodes and links,
A display method of a change impact support apparatus, comprising: a display control step of distinguishing the display form of the node or link according to the magnitude or occurrence rate of the change and outputting to the display device.
  The display control step calculates the expected value of impact to the end of the tree using the impact and occurrence rate of the additional man-hours or additional costs from the upstream of the tree, and displays the expected value corresponding to the node at the end of the tree. The display method of the change impact support apparatus according to claim 15, wherein:

Images