Skip to main content

Fujitsu

Japan

Archived content

NOTE: this is an archived page and the content is likely to be out of date.

Abstracts of Magazine FUJITSU 2009-11 (VOL. 60, NO. 6)

Special Issue: Fujitsu's Activities for Systems Engineers' Manufacturing Innovation

  • New Framework for Systems Engineers' Manufacturing Innovation

To ensure that systems engineers maintain their high technical capabilities for addressing the significant trends toward openness and various changes surrounding them, the Fujitsu Group has been working on the Four Innovations activities: Design, Production, Maintenance, and Way of Working for systems engineers. The results of our activities conducted for two and a half years include requirements definition guidelines and basic design guidelines and the establishment of a framework of maintenance by Application Portfolio Management (APM). At the same time, these activities have clarified challenges as well. To solve these challenges, we will carry the Four Innovations forward to the second stage, and add Human Resource Innovations to provide Five Innovations, which we will advance in the next three years. However, the Five Innovations are not the final target. We will consolidate Design, Production, and Maintenance into a system (knowledge) for offering value to customers in an integrated form. In addition, the Way of Working for systems engineers and human resources will develop next-generation systems engineers with deep knowledge. The goal is to establish a structure to allow knowledge and human resources to be synchronized so as to continuously create value.

  • Consensus Building by Structured Requirements

The importance of upstream processes in system development has been recognized for a long time. However, the failures that occur during the requirement definition process show no sign of decreasing. This is partly because the process of requirement definition itself has become more difficult considering the times we are in and the environmental changes taking place. However, there is also a more fundamental problem with the current requirement definition process.
This paper introduces our approach to achieving both (1) consensus building based on a layered structure of interested parties and continuity of demands, and (2) traceability (i.e., a structured requirement management technique).

  • New Technique for Estimating Development Scale—Implementing the Function Scale Method—

The Function Scale (FS) method was developed by Fujitsu in 2004, and since then it has been extensively used in the Fujitsu Group. Compared with the Function Point (FP) method, which is the de facto standard for SLOC estimates and the measurement technique for measuring function scale that has been generally used up to now, the FS method aims to produce the same results no matter who it is used by as it is easy to understand and easy to use. This paper explains that the FS method is not merely a measuring technique and gives actual examples of how it can be used in project management. Moreover, this paper describes the various measures that are being implemented to promote the spread of the FS method within the Fujitsu Group.

  • New Approach to the Verification of Application Software Quality

As part of the reformation of production, this paper introduces a new approach that aims to greatly improve application software quality by using formal verification technology. Our technology automates test scenario and data generation, executes tests, and confirms the results by formally describing the external specifications of applications. As a result, major problems with the current test method can be resolved, such as insufficient consideration being given to test cases and test data, too many test cases to execute, human errors when executing tests and confirming the results, and discrepancies arising from a dependency on individual testers. Moreover, by combining our technology with an application framework, the parts of application programs that need to be verified can be narrowed, making the technology applicable to practical-scale applications. We will proceed with research towards the practical use of this technology, by for example expanding the range of applications that can be verified with it.

  • Process-oriented Approach to Industrialization of Software Development

Fujitsu Applications, Ltd. was founded in April 2004 as a development division based on the System Development Architecture & Support facilities (abbreviated as SDAS, these are Fujitsu's comprehensive system development methodology), with the mission of implementing SDAS and establishing and accumulating development know-how. The company is working on new technologies by conducting trial runs on separation of design and manufacturing, industrialization of software development, and use of offshore development. In the Fujitsu Group, SDAS plays the roles of manufacturing and testing plants, and it is promoting collaboration with the field in manufacturing innovation including technological development and verification and establishment of development management methodology from the viewpoint of the field. This paper describes the approach to the industrialization of software development, seen as a process of improvement from a personal development style to a process-oriented systematic development style. In addition, this paper presents the code of conduct of autonomous improvement activities called the "Four Rules" and "Six Mechanisms," together with specific data on achievements.

  • Improving Project Quality by Applying Third Party Quality Verification

Third party quality verification is a tool to help improve the quality of projects, such as system integration projects. Fujitsu Advanced Quality Limited is now working on third party quality verification in the production process of SI projects. The main features of our approach are to pay attention not to products but to people (work quality) and focus not on managers but on persons actually working in the field as the verification targets. According to principle of the Three A's (where the A's stand for actual things), we go to sites in the field, interview the persons actually working there while checking actual things (review records etc.), and verify the actual reality (i.e., work quality). In addition, we feed the verification outcome back to the project with our suggestions for improvement. As a result, many work quality issues can be resolved, and we can thereby improve the project quality. This paper introduces the approach of Fujitsu Advanced Quality Limited to third party quality verification, and gives some examples of its practice.

  • Strengthening Service Management by the Comprehensive Analysis of Incidents

The complexity of the information systems (business applications and infrastructure) that support business enterprises has been increasing year on year, and in accordance with this the difficulty of operating and maintaining them has also been increasing. Daily incidents need to be dealt with in the places where these systems are operated, and there are not many projects for which the overall tendencies of those incidents can be determined, and the stage of solving their root cause or making proposals for improvement is seldom reached. To support the smooth operation of business, the APM (Application Portfolio Management) service center in Fujitsu has established a technique for comprehensive incident analysis, such as those incidents arising from inquiries and messages from the infrastructure and applications, from the perspectives of service management, users, and business operations. This technique targets busy fields in the operation and maintenance process and is composed of a service and templates that ease the burden placed on service managers in terms of time and effort, helping them find overall tendencies and root causes. This paper describes the technique for comprehensive incident analysis, and examines the effects of that technique by looking at actual examples.

  • Expansion of Application Configuration Management

Application maintenance requires total success when changing existing applications. Even when changing a few lines of programming, it is necessary to investigate the impact of those changes and conduct tests to make sure that they do not affect other configuration items. These investigations and tests make up 80% to 90% of the application maintenance cost. The accuracy and efficiency of investigating the effects of any change need to be improved. Usual investigation depends on experienced and skilled engineers who are in charge of the system and application. They not only investigate applications but also the framework of applications and infrastructure. They also consider operations. We arranged configuration items that are needed for investigations and constructed a model. That model was applied to CentraSite's registry, and we made an investigation method. By using that method, all the engineers working in the operation and maintenance process can conduct investigations with the same accuracy and efficiency.

  • Innovation of Working Style through Ethnography for Business and Organizational Monitor

Understanding the nature of clients' work is a precondition for helping them to solve challenges posed by their business practices. Fujitsu has been committed to developing and deploying Ethnography for Business that merges ethnographic qualitative analysis, frequently used in the areas of marketing and product design, with conventional quantitative analysis, not for study but as business. While ethnography has been used in the area of research, our current approach utilizes this methodology to understand the nature of clients'work. Also, in the field of research on organizations which has become widely recognized in some international ethnography associations including EPIC, Fujitsu's approach of Organizational Monitor has been developed and implemented by integrating the factors of quantitative analysis emphasizing the continuous and proactive commitments to innovation by organizations and individuals. It is regarded as an approach that goes beyond the conventional academic approaches that focus only on the reporting of exploration and interpretation of the facts observed in business scenes. This paper introduces a new approach based on these two types of ethnographic technique that adds a new aspect to the conventional working style.

  • New Efforts for ProjectWEB to Be a Business Infrastructure

Since its introduction in 1998, ProjectWEB, Fujitsu's in-house knowledge management tool has played an essential role in Fujitsu's system engineering work. Recently, increasing numbers of business units have been using ProjectWEB to prevent security incidents, which demonstrates its ability to be used not only in projects, but also in various other business arenas. Not content with its current role, we will aim to position ProjectWEB as a business infrastructure for all Fujitsu engineers to be used also outside the Company. In order to achieve this, three priority subjects have been set: 1) realization of multiple-project management, 2) promotion of rules and manners, and 3) improvement of service levels.
This paper will introduce the background to why ProjectWEB has been developed, its major features, and the approaches taken toward the goal of making it a business infrastructure.

  • Innovating Talent in Solution Business

The evolution of technology has accelerated changes in business domains and contents, so that there is now intensified competition to acquire new markets. In this situation, companies need to concentrate on their core competences. They must create new business domains voluntarily, by making the most of the contents they have. We are in an age where it is difficult to survive in business, and companies should grow via their own unique strategies. Fujitsu is providing solutions, mainly technological ones, to accelerate the expansion of our global business in both circles of services and products. We are developing talented people who can provide value and satisfaction for our customers. This paper introduces the concept behind and course of actions for this talent innovation, which we are conducting to master advanced technologies appropriately to fit the purpose.

  • Attempting to Increase Longevity of Applications Based on New SaaS/Cloud Technology

In recent years, SaaS/Cloud technology has advanced significantly in development and operation, and IT vendors including Fujitsu must keep up with these developments. In this field, we need to ensure applications have a long life cycle. This paper introduces the concept of a meta-framework, which is a framework for resolving the issues involved in extending the life of applications. The basic idea of a meta-framework is to validate the design and development process, and separate the design and framework processes. This will bring us closer to the ideal form of software development in which the end of design signifies the end of testing. By separating the design and the framework, the design can be a permanent asset. In addition, by creating a common base, i.e. the meta-framework, it will be possible to easily migrate a system to the latest framework. We will be able to create templates that are used to automatically generate source codes, and expanding the scope of automatic generation will reduce the cost of migration.