Method and system for ensuring completeness and quality of a multi-component project

[0002] The present invention relates generally to computer systems, and more particularly to computer systems that ensure completeness and quality of a project.

[0003] The human memory is imperfect. Most people are unable to remember everything they have learned. And, even if a particular person has an especially good memory, he may nevertheless be unable to quickly and to accurately retrieve all relevant information related to a particular project. In view of the foregoing, people often rely on various devices to help retain and retrieve knowledge. For example, a baker uses recipes to help him bake a variety of bread, cookies, and cakes. Each recipe contains a list of ingredients and a set of baking procedures. By using these recipes, the baker can be sure that he will not leave out an ingredient or overcook the food, thereby producing baked goods of consistent quality.

[0004] A more sophisticated example of a memory assistance device is a checklist typically used by automobile insurance claim specialists. There are many types of vehicles in the market today, and each type of vehicle has different models and each model has thousands of parts. Relying on memory alone, it would be virtually impossible for a claim specialist to provide a reasonably accurate “parts and labor” estimate for a particular vehicle model. Thus, the claim specialist needs a checklist that contains, among other things, a list of parts which require replacement, parts costs, and time periods to install those parts. With the checklist, even a relatively inexperienced claim specialist can provide a reasonably accurate parts and labor estimate.

[0005] While a recipe is helpful to a baker and a checklist to an insurance claims specialist, such devices are inadequate for ensuring completeness and quality of a more complex project that requires involvement of many people of different skills. For example, the above-described memory devices are inadequate to assist an automobile manufacturer who relies on a team of mechanical engineers, electronics engineers, designers, metallurgists, materials scientists, psychologists, marketers and lawyers, to design and build an automobile that performs as desired, sells, and meets regulatory requirements. It is clearly impossible for the automobile manufacturer to rely on human memories alone to determine how, for example, changes made to the fuel system of an automobile might affect the exhaust system, the emission system, and the ignition system of that automobile. This “memory” problem is exacerbated since it is not uncommon for skilled employees to leave the company thereby losing individual and institutional knowledge.

[0006] Consider a very simple scenario in which the automobile manufacturer plans to redesign the fuel system of one of its automobile models. The automobile manufacturer needs to know what factors must be considered and how many hours will be required to redesign the fuel system for that particular automobile model. The first group of people the automobile manufacturer likely consults for implementing the redesign is the fuel system engineers group. In addition to providing the time estimate and factors to consider, the fuel system engineers group, under ideal circumstances, notifies the automobile manufacturer that it (the manufacturer) should also consult engineers from the emission, exhaust, ignition, and power output groups because changes made to a fuel system typically require redesigning these other systems as well. The automobile manufacturer preferably then consults the other engineering groups as suggested. If each engineer of each team properly remembers both the necessary actions and the time required to implement those actions to redesign each system, the automobile manufacturer will obtain a reasonably accurate project description and time estimate. But the reality is that one or more “key” engineers may no longer be with the company, or those with the company may provide inaccurate descriptions of “necessary actions” or time estimates thereby leading to poor project management plans. Further, one or more of the engineers might employ in work descriptions acronyms and ambiguous words, making it difficult for the automobile manufacturer to accurately predict the true scope of a proposed design change.

[0007] Another example of a complex system where a change to one component might require changes to other components is a complex billing system, and more particularly, a computerized billing system of a telephone company. As mobile telephone systems, data lines and other services provided by the telephone company have become more common, billing individual customers for such services has become increasingly complex and requires hundreds of programmers and managers to ensure efficient and accurate billing.

[0008] Existing software products, such as CBR Express, available from Inference Corporation of Novato, Calif., which are directed primarily to individual case retrieval do not provide the aforementioned functionality as they are not designed to enumerate and analyze all combinations or permutations of all potential impacts of a project.

[0009] There are also other products that are related to document version management, such as DOORS , available from Quality Systems & Software of Mt. Arlington, N.J. These products assign version numbers, e.g., “1.0” to a first document and when subsequent changes are made to that document, the version is automatically updated to “1.1,” “1.2,” etc., along with annotations and dates with respect to who made the changes and why. However, such products also are unable to accomplish the tasks enumerated above.

[0010] Also known are “expert systems.” These systems typically are designed as either rule-based or case-based. With either type, however, it is necessary to program all of the rules and/or all of the possible case scenarios that might be encountered. Unfortunately, it is not always clear, at a given time, what changes or modifications to a system might be effected at a future time, thereby defeating the usefulness of both types of systems. Moreover, these systems require extensive programming and maintenance.

[0011] The Requirements Expert of the present invention fills the need for an experienced, meticulous, and detail-oriented analyst. The Requirements Expert is a dynamic, flexible tool, which uses a knowledge base that is created by one or more users. The Requirements Expert gains additional knowledge each time it is used. The Requirements Expert helps its user to determine what the scope or requirements are for a particular project. It also gives its user an estimate of time required to complete the project. The Requirements Expert further ensures that its user considers every factor associated with that project by systematically asking the user a series of questions.

[0012] In a preferred embodiment, the Requirements Expert (“R.E.”) is implemented in a computer system which has a central processing unit (“CPU”), a memory and input (e.g., keyboard) and output (e.g., monitor and/or printer) systems. The computer system is programmed with a relational database and algorithm which facilitates entry, manipulation and retrieval of information from the database. The database can be implemented in any general purpose relational database application such as MICROSOFT ACCESS™ and the algorithm can be written in any suitable application such as VISUAL BASIC™. The information stored in the database is knowledge regarding a particular project or complex system. The complex system preferably is represented as a logical model consisting of nodes and links where each node represents a particular component of the system and the links connecting the nodes represent relationships between any two nodes. Each node, i.e., system component, is further supplied with a list of questions pertaining to that node and information with respect to how that node might affect other nodes. The output of the Requirements Expert preferably is used to provide a listing or printed document of the analysis performed by the Requirements Expert.

[0013] Before using the Requirements Expert, a user preferably first provides the Requirements Expert with information about each system component and how changes to each component might impact other components. This information preferably is entered in the form of questions to be directed to a user. Once the nodes and links are defined, and at least some questions for each node are stored in the database, the Requirements Expert is ready for use. When the algorithm is executed, the Requirements Expert queries the user to identify which of the nodes is to be considered to analyze a particular project. Upon obtaining the response, the Requirements Expert asks the questions associated with that node, and provides the user with other relevant information about that node, such as how long it might take to accomplish a proposed change. The Requirements Expert also asks questions, or presents queries, associated with related effected nodes, consistent with the logical model represented in the database. The Requirements Expert also records, or stores, responses to the questions along with definitions for any detected acronyms or ambiguous words. Finally, the Requirements Expert provides the user with a document listing the questions, answers to the questions, and, as available, a listing of recommended actions to be taken to accomplish the proposed change. Other relevant information, such as the time associated with accomplishing the proposed change is also provided, as available.

[0014] Thus, in the preferred embodiment of the present invention the Requirements Expert has the following characteristics or objectives:

[0015] 1. It helps its user to retain and retrieve knowledge.

[0016] 2. It ensures that the user considers every detail of a project or complex system.

[0017] 3. It assists a company in retaining its institutional expertise even when key employees leave the company.

[0018] 4. It describes to the user the nature of each component as that component is encountered.

[0019] 5. It informs the user how one component or portion of the complex system impacts other components or portions.

[0020] 6. It queries the user with a series of questions about each component or portion of the system and records answers to those questions.

[0021] 7. It detects acronyms and requires the user to provide a definition for each acronym detected.

[0022] 8. It detects ambiguous words and requires the user to replace each ambiguous word with a more definite term or phrase.

[0023] 9. It provides a time estimate for performing the project.

[0024] 10. It provides an output document, free of ambiguous words and undefined acronyms, that lists each impacted component, list of questions and answers to the questions about each component, and an estimate of the time required to complete the project.

[0025] The Requirements Expert is also unique in that its emphasis is on requirements acquisition, not only requirements management. That is, the Requirements Expert of the present invention assists the user to develop new action or project plans, including accurate projected time for carrying out those plans. This is different from conventional requirements management where a system or person is responsible for ensuring that predetermined requirements are met.

[0026] FIG. 1 is a schematic diagram illustrating the various elements of the Requirements Expert of the present invention.

[0027] FIG. 2 is a flowchart illustrating the steps required to provide initial knowledge to the Requirements Expert.

[0028] FIG. 2A is a schematic diagram of an example of the present invention, representing a model of how various systems of an automobile relate to each other.

[0029] FIG. 3 is a flowchart illustrating the steps required to use the Requirements Expert after it has been provided with initial knowledge.

[0030] FIG. 4 is a schematic diagram of another example of the present invention, representing a model of how various components of a billing system relate to each other.

[0031] FIGS. 5A-5B illustrate an exemplary listing of nodes representing the various components of a complex billing system.

[0032] FIGS. 6A-6B illustrate an exemplary listing of relationships between nodes illustrated in FIGS. 5A-5B.

[0033] FIGS. 7A-7E depict an exemplary listing of questions associated with selected nodes.

[0034] FIGS. 8A-8E depict a partial example of an output of the present invention including questions and time periods.

[0035] The present invention, the Requirements Expert, is a computer system that ensures completeness and quality of a project.

[0036] FIG. 1 is a schematic diagram illustrating the various elements of the Requirements Expert (“R.E.”) of the present invention. As shown in FIG. 1, the R.E. has an algorithm 10 and a knowledge base 11 stored in a memory 12, a central processing unit (“CPU”) 13, an input device 14, and an output device 15. Input device 14 (such as a keyboard) and output device 15 (such as a monitor or a printer) are in communication with CPU 13 as in conventional computer systems. Memory 12, within which algorithm 10 and knowledge base 11 reside, is also in communication with CPU 13.

[0037] Knowledge base 11 preferably is manually constructed by human experts such as engineers, scientists, marketers and other experts. Execution of algorithm 10 prompts a user to update knowledge base 11, and any information stored in knowledge base 11 is retrievable by algorithm 10. In the preferred embodiment, algorithm 10 is written in the VISUAL BASIC™ programming language and knowledge base 11 is implemented in a relational database such as MICROSOFT ACCESS™. Of course, any other suitable application for algorithm 10 or knowledge base 11 is considered within the scope of the present invention. For example, an ORACLE™ relational database may be utilized to construct knowledge base 11.

[0038] Knowledge base 11 is initially empty. Then, a logical model of a complex system is created in knowledge base 11 through the execution of algorithm 10 by using input device 14. Conceptually, the logical model is a network of nodes and links in which nodes represent domain items, i.e., components of the complex system, and links represent node-to-node impacts, i.e., relationships between nodes or system components. As the user continues to use the Requirements Expert, the user incrementally builds knowledge base 11, and the Requirements Expert becomes more useful to the user over time.

[0039] FIG. 2 is a flowchart illustrating the preferred steps to provide initial knowledge to the Requirements Expert. In step 21, a Requirements Expert user creates a logical model for a complex system. FIG. 2A is a schematic diagram of an example of a logical model 200 illustrating how various components or systems of an automobile relate to each other. Consider a scenario in which an automobile manufacturer has a car model that has been in production. The automobile manufacturer thereafter decides to incorporate new features in the car, and the automobile manufacturer needs to know what actions and/or how much time are/is required to redesign the car.

[0040] As shown in FIG. 2A, logical model 200 represents the following components: a fuel system 201, a power output system 202, an exhaust system 203, an emission system 204, a cooling system 205, and an ignition system 206. The arrows connecting the various systems represent the “parent-child” relationships or links among the various systems. For example, changes to fuel system 201 affect power output system 202, exhaust system 203, emission system 204, and ignition system 206. However, changes to fuel system 201 do not affect cooling system 205. Thus, each of the systems or components of the overall complex system is considered a “node,” and each arrow represents a “link.”

[0041] Referring back to FIG. 2, we now consider the automobile manufacturer's project as an example to describe the remaining steps in FIG. 2. The automobile manufacturer, in step 22, defines each node by creating a table in knowledge base 11. Table 1 is an example of node definitions for the automobile project summarizing the names of the various components, a brief description for each component, and how many hours are required to redesign various aspects of each component.

[0042] In step 23, the automobile manufacturer defines each link consistent with logical model 200 (shown in FIG. 2A). Each link connects a “parent” node to a “child” node. The parent-child relationship is one in which changes made to a parent node tends to cause a recommended modification to a child node. Table 2 (which is stored in knowledge base 11) represents all the parent-child relationships in the automobile example.

[0043] As mentioned above, a link between any two nodes is a parent-child relationship. Once all parent-child relationships have been defined in knowledge base 11, the Requirements Expert is ready to assist the user in making changes to a complex system.

[0044] In step 24, the automobile manufacturer supplies knowledge base 11 with a list of questions for each node of model 200. By repeating steps 24 and 25, Tables 3A through 3F are created in knowledge base 11.

[0045] FIG. 3 is a flowchart illustrating the preferred steps for using the Requirements Expert (abbreviated as “R.E.” in FIG. 3) after it has been provided with initial knowledge as described above. For example, when the automobile manufacturer executes algorithm 10, the Requirements Expert, in step 31, queries the user, that is, asks the user to identify which of the six sub-systems or components (nodes) is to be considered for implementing a project, i.e., effecting a change to the overall system. Suppose the automobile manufacturer wants to have an estimate of how long it might take to redesign the fuel system. The automobile manufacturer would provide “Fuel System” as a response. Upon encountering the input “Fuel System,” the Requirements Expert searches Table 1 and finds fuel system 201 in model 200. The Requirements Expert then informs the user that the “Fuel System” relates to carburetor and fuel injection, as described in Table 1. The Requirements Expert then proceeds to step 32 and asks the questions listed in Table 3D, which are associated with the fuel system node. The purpose of these queries is to ensure that the user is aware of every detail associated with fuel system 201 and every way in which changes to fuel system 201 might impact other systems. Also in step 32, the Requirements Expert records the answers received. When the Requirements Expert has recorded the answers, it informs the user, in step 33, that it takes 20 hours to redesign fuel system 201 as shown in Table 1. In step 34, the Requirements Expert adds the 20 hours to the project total.

[0046] The Requirements Expert then checks Table 2, in step 35, and determines that emission system 204 is a child node of fuel system 201. In step 36, the Requirements Expert asks the user whether to consider the questions associated with emission system 204. If the answer is “Yes,” the Requirements Expert repeats steps 32 through 34 and asks all the questions in Table 3B, and adds the 50 hours from emission system 204 to the project total resulting in a new total of 70 hours. Summing is performed by a summing circuit (not shown). If the user considers every child node associated with fuel system 201, the project total would be 150 hours (20 hours from fuel system 201, 50 hours from emission system 202, 30 hours from exhaust system 203, 40 hours from ignition system 205, and 10 hours from power system 206). As shown in FIG. 3, the user has the option of bypassing any of the child nodes in step 36.

[0047] In step 37, the Requirements Expert asks the user whether the user wishes to consider another parent node. For example, the user may also want to redesign cooling system 205, in which case the process will repeat steps 32 through 37 and the new project total will be 210 hours (150 plus 60 from cooling system 205). The Requirements Expert preferably then prints, in step 38, a document that includes the questions and answers, and the total hours required for the project.

[0048] Again, as shown in FIG. 3, it is possible for the user to bypass any child node to streamline the process, especially if a particular child node has already been identified via a different parent node. Further, as is evident from all of the foregoing, the Requirements Expert of the present invention keeps both a listing of recommended actions and a running total of time for completing those actions. Accordingly, the Requirements Expert provides an especially useful tool for project planners and managers.

[0049] In the preferred embodiment, the Requirements Expert also detects acronyms and ambiguous words in steps 31 and 32. When the Requirements Expert encounters an acronym for the first time, it asks the user, via a query engine, what the acronym stands for, and keeps that definition in knowledge base 11. For example, if the user had supplied the acronym “EFI” in one of the responses, the Requirements Expert checks an acronym list in knowledge base 11 to determine whether “EFI” has a definition. If it does not, the Requirements Expert asks the user to supply a definition, e.g., “electronic fuel injection,” and adds the definition to the acronym list. As another example, the automobile manufacturer may define the acronym “HP” as “horse power.” Table 4, stored in knowledge base 11, is an example of an acronym list.

[0050] The second time the Requirements Expert encounters the term “HP,” it looks up Table 4 and asks the user, “Does HP mean Horse Power?” If it does, the process continues with the next process step. Otherwise, the user supplies the second meaning, e.g., “HEWLETT-PACKARD™,” and the Requirements Expert updates Table 4 to become Table 4a below:

[0051] When the Requirements Expert encounters the term “HP” for a third time, it asks the user, “There are two meanings for HP. Which do you mean, Horse Power or HEWLETT-PACKARD™?” If there is yet a third definition for HP, then the Requirements Expert allows the user to input the third definition so that the next time the user inputs “HP” the Requirements Expert queries the user to choose from a list of three possible meanings associated with “HP.”

[0052] Similarly, if the user provides an ambiguous word that is listed on the Requirements Expert's list of ambiguous words, it prompts the user to provide the proper meaning of the ambiguous word. Table 5 is an example of a list of ambiguous words.

[0053] In view of the potential for a great number of nodes that might require consideration for a given project, the Requirements Expert preferably also includes a means of remembering the state of a user's interaction with the Requirements Expert and returning the user to that state following an interruption, either intentional or accidental, of the user's Requirements Expert session. The means for remembering may be implemented with an automatic periodic saving of a session including the questions and answers to the questions that have already been entered. The in-progress information may be saved to a disk or network.

[0054] Another example of a complex system that can benefit from the Requirements Expert of the present invention is a complex computerized billing system associated with a telephone company. FIG. 4 is a sub-system or component logical model 400 for such a billing system. In FIG. 4, element 401 is used for general information. While this node does not represent a component of the system model, per se, general information is desirable and is representative of generic questions/information. For example, node 401 preferably includes descriptions of proposed changes, or the reason for the proposed change or the requested date of completion. In this case, the arrow from node 401 to node 402 does not reflect an “impact” in the sense of the present invention, but in this instance is used only to illustrate a direction of flow from node-to-node.

[0055] Node 402 represents rating, which refers to the rates charged for specific types of telephone calls. Factors affecting a rating might include time of day, originating point, terminating point, etc. Node 403 represents service (abbreviated as “SVC” in FIG. 4) rating, which refers to the type of service used, such as flat rate local service, or measured local service.

[0056] Node 404 is for recurring charges which occur on each and every monthly telephone bill. Examples of such recurring charges are for caller ID and voice messaging services. Node 405 represents service orders, which are customer orders for new products and services. Node 406 models one time charges, such as installation charges. Node 407 covers message (abbreviated as “MSG” in FIG. 4) rating, which is directed primarily to how long distance calls are billed.

[0057] Node 408 refers to bulk rate services such as WATS lines, etc. Node 409 represents those charges which are specific to the local Bell Operating Company (abbreviated as “BOC” in FIG. 4). Node 410 is concerned with a telephone service which is billed at a flat rate (i.e., which is not dependent on amount of service). Node 411 represents aspects of local telephone calls, as opposed to long distance. In this example, all of the nodes, except node 401, represent segments of an overall computer program that implements the billing system. Thus, in view of this example and the automobile manufacturing example discussed previously, it is evident that the Requirements Expert of the present invention is applicable to both complex mechanical and software systems.

[0058] FIGS. 5A and 5B are an exemplary listing of nodes representing the various components of a complex billing system.

[0059] FIGS. 6A and 6B illustrate an exemplary listing of the parent-child relationships between the nodes illustrated in FIGS. 5A and 5B. This listing is a typical printout of the Requirements Expert knowledge base 11, if asked to output all nodes and their relationships.

[0060] FIGS. 7A through 7E illustrate an exemplary listing of questions that might be associated with selected nodes of FIG. 4. As can be seen from FIGS. 7A-7E, the extent and detail of questions can be designed to suit the particular circumstances. Finally, FIGS. 8A through 8E depict a partial example of an output of the present invention including questions, whether a particular question was answered and a time period associated with completing an action with respect to modifying that aspect of the complex system.

[0061] Thus, the Requirements Expert of the present invention is particularly useful in an environment in which complex systems are maintained by a plurality of people. The Requirements Expert stores the knowledge of those plurality of people in a central database and the database can be accessed by a single user. The Requirements Expert thus gives that user the capability to develop a clear list of actions to be taken to complete a project or implement a change to a complex system. In accordance with the present invention, the user is provided with an impact analysis, i.e., which components of the system are impacted by a proposed change, an assurance that all details having to do with a change have been properly addressed, and a clear description of what work or action must be taken and how long that action or work is likely to take.

[0062] More particularly, the Requirements Expert of the present invention formalizes and automates a procedure which is typically informal, manual, incomplete and error-prone. Impact analysis, especially in the software field, is typically done by programmers who “just know” and have personal long-term working relationships with each other. Obviously this is a source of mistakes and omissions. Further, it is generally assumed that everyone in an organization understands relevant acronyms and commonly-used terminology. Likewise, estimates of time required for project completion is often an informal and imprecise process. In contrast, the Requirements Expert establishes a formal, automatic, complete review system for a project, thereby substantially minimizing errors. Thus, the Requirements Expert makes the process of maintaining existing software or other complex systems a more defined, repeatable process, so that when changes are implemented and the new product is released there are fewer mistakes and fewer instances in which this new product or system has to be retooled.

[0063] The Requirements Expert of the present invention may also be considered to be a requirements generating apparatus for establishing requirements for implementing changes to a system. That is, the Requirements Expert includes a relational database programmed to operate as a network of interconnected nodes, wherein each of the nodes is capable of storing query information related to a distinct component of the system. Also included is a query engine associated with the relational database. The query engine queries a user based on stored query information. Finally, to provide the total time associated with implementing a particular project, a summing circuit is provided that is in communication with the relational database and query engine. The summing circuit sums periods of time associated with specific actions for implementing the changes to the system, wherein the specific actions are determined by responses received by the query engine. The Requirements Expert of the present invention, implemented as a requirements generating apparatus, can be implemented with any general purpose computer system or specially designed computer system having the elements depicted in FIG. 1.

[0064] The foregoing disclosure of embodiments of the present invention and specific examples illustrating the present invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many variations and modifications of the embodiments described herein will be obvious to one of ordinary skill in the art in light of the above disclosure. The scope of the invention is to be defined only by the claims appended hereto, and by their equivalents.