There is no need to dwell upon the great contribution of science and technology to human welfare. Research activities are important to the advancement and promotion of science and technology. Various organizations have been established for the purpose of performing research activities.

The first problem that we encounter is the method of selecting research projects; careful consideration of "what and how" on research projects is precondition in performing effective research works.

The function of selecting research projects is closely related with the method of evaluating the projects. Then, the expression of "pre-evaluation" seems to be more suitable than "selection" for the present purpose.

The difficulties of evaluating research projects comprise of "difficulties of establishing proper standard" and of "difficulties of motivating research members".

A research management based on a uniform evaluation standard and a concept of contract would bring about a doubtful efficiency of research activities, as well as that without a clear evaluation system in the excess of stressing motivation of research members.

The authors intend to discuss in the present paper neither "an absolute scaling for evaluation" nor "a uniform method of motivating research members". We have a view point that both the value scaling and the motivation should finally be transfered to some relative ones when they are used in an actual decision-making procedure. Therefore, we aim at proposing merely a tentative evaluation system which may be used to obtain an potimum solution by repeating "try and error processes".

The authors will be grateful if the present paper could make any contribution toward the field of research management.

* Manuscript received on January 23, 1968.
** Member of Research Planning and Infomation Division, Building Research Institute, Ministry of Construction