This paper discusses a design and implementation method of a solver, which is a main engine for a problem solving type management information system. In particular, a systems approach is applied to the method.

The systems approach of this paper indicates modeling in the set theory using system concepts of the general systems theory and implements the model in the extProlog, whish is a Prolog extended for implementation of a management information system.

A development procedure is proposed for the design and implementation, which consists of 6 stages. Problems are classified into 8 categories for the sake of effective application of the procedure. Modeling is discussed according to the classification.

In order to clarify arguments simple examples are used to illustrate concepts and procedures.

In Appendixes real programs, which are implementations of the method for the examples, are presented. Finally, the paper presents a formalization of the solver.

In order to draw up the information strategy, the first task is to identify important jobs in the execution of the management strategy that can be effectively supported by information technologies. This paper shows a method to perform such an undertaking by giving a case study. This method begins by following the balanced scorecard approach whose purpose is to evaluate the corporate vision and management strategy from the financial, customer, internal business processes, and learning/growth perspectives. Following this approach, we draw the strategy map to clarify the strategic objectives and then the details of their evaluation indicators and causal relationship. Then we draw the information birds-eye view that is proposed in this paper to indicate the relation between the management and information strategies. Utilizing these figures, management, strategy planning departments, information system departments, and users of information systems can understand the important jobs that can be effectively supported by information technologies, and their causal relationship among the four perspectives.

We proposed a design and implementation approach of a problem solving system. This approach designs a system using concepts of GST (general systems theory) described in the set theoretic terms and implements the designed system in the extProlog (extended Prolog). This approach is called MGST (mathematical GST) approach. The approach is proposed to fill the gap between a solver image and its realized system.

In the approach a problem solving system is constructed as composed of two components, user model and goal seeker. In order to make the approach be easily used by less technically oriented people, the goal seeker is built as a standardized component so that a novice user has only to produce a user model and attach it to the standardized component to get a problem solving system.

This paper will discuss a data mining system as a non-trivial application of the approach. It will be demonstrated that a data mining system is designed as a problem solving system following the approach.

The financial institution of our country was homogeneous and standardized corporate strategy for every business category called "convoy system", but the financial institution competition came to be performed among more companies. In the financial institution, practical use of the marketing strategy that utilized IT (Information Technology) positively became important by progress of electronic society. In this writing, it does not surveys on "Push Strategy" and "Pull Strategy" which are a promotion strategy of a manufacturing industry in the marketing strategy of a financial institution, and also the policy of "Data Base Marketing" and " Account Aggregation Service " which is the marketing strategy that applied the framework and IT of a promotion strategy is presented and the possibility of the marketing strategy is considered.

It is expected that implementation of telework will improve office workers productivity through concentration on their work and improved worker’s autonomy. A survey was conducted to investigate the factors that influence the productivity improvement. As a result, a multiple regression model, which consists of four explanatory factors to explain performance of teleworkers, was obtained. The four factors are: actively getting knowledge and experimental information; frequent communication between managers and workers; access to data for their routine work; and an incentive by a proper evaluation for telework. It is confirmed that the multiple regression model could explain the performance of teleworkers very well.