###### Abstract

▼

To provide a fundamental data for productivity improvement planning and a criterion toward incentive pay allocation among organizational units for top management, it is important to identify factors affecting military R&D productivity and to develop productivity measurement model as well as producti...

To provide a fundamental data for productivity improvement planning and a criterion toward incentive pay allocation among organizational units for top management, it is important to identify factors affecting military R&D productivity and to develop productivity measurement model as well as productivity evaluation model reflecting characteristics among organizational units. These needs lead to this study to tackle those problems and the main results of this study are threefold. The first part identifies the effect factors on military R&D productivity. To discover effect factors, questionnaire, which includes 51 items related to productivity, was mailed to 131 R&D team leaders of Agency for Defense Development (ADD) that has five R&D centers and three support centers, asking for rating each item with five-point answer scale. The total number of questionnaires answered was 117 (89.3% of response rate), but 10 questionnaires were not usable because of the missing data. This study statistically analyzes the returned questionnaires to identify factors affecting productivity and to find the most important factor of them to each of five R&D centers. To find effect factors, using SAS program, factor analysis has been performed to 51 items that are shown in the appendix. According to the results of factor analysis, this study determined the following 8 factors: (i) research climate, (ii) quality of R&D management, (iii) research environment, (iv) ease of communication, (v) development of professional ability, (vi) cooperative relations within organization, (vii) satisfaction in R&D activity, and (viii) motivation of researchers. Using multiple linear regression analysis, where the factors discovered become eight Independent variables and the level of R&D productivity stated in questionnaire become the dependent variable, this study determines the most important factor affecting to productivity improvement for each of five R&D centers. It includes the following: (i) cooperative relations within organization for the ground weapon systems center, (ii) quality of R&D management for the naval weapon systems center, (iii) satisfaction in R&D activity for the aircraft systems center, (iv) motivation of researchers for the missile systems center, and (v) research environment for the communications and electronics systems center. Thus, it is necessary for each of five military R&D centers to establish more focused productivity improvement planning regarding the above factors. The second part develops productivity measurement model as well as productivity evaluation one taking into account characteristics among five military R&D centers. The proposed models that employ a partial factor productivity measure with one input factor and some output factors, offer a valid tool for productivity measurement and evaluation of an R&D organization. The partial factor productivity measure is obtained by dividing one input factor into the weighted sum of each output factor. In particular, the productivity evaluation model is an adjusted static or dynamic index with regression equations for productivity measure through characteristic variables of an R&D organization. Using this model, the results can provide a criterion of allocating incentive payment among organizational units for top management. The third part presents an illustrative example, which applies the proposed models to five Military R&D centers and demonstrates their applicability. To apply these models to a military R&D organization for their usefulness, this study included the following: 1) To identify three types of input factors, namely, the personnel engaged in an R&D center, the labor cost paid to the personnel engaged in an R&D center, and the total expenditure of an R&D center. 2) To identify output factors and their sub-output factors through questionnaire. The list of 7 output factors shows technical report (TR), award (AD), software (SW), hardware (HW), journal article (JA), conference presentation (CP), and patent (PT). The list of 18 sub-output factors includes technical report, gold award, silver award, bronze award, and so on. 3) To determine the weights of output factors and their sub-output factors using questionnaires and Analytic Hierarchy Process (AHP) software. The relative weights of each output factor are TR: 0.085, AD: 0.103, SW: 0.195, HW: 0.236, JA: 0.101, CP: 0.061, and PT: 0.219. 4) To classify productivity measures, which consist of 3 classes of partial factor productivity measures with 3 types of input factors. 5) To identify measurable characteristic variables of five military R&D centers through interview and statistical analysis. The 26 measurable characteristic variables are grouped into three areas by their nature. The first area is the investment type, which consists of 9 variables closely related to the pattern of resource allocation. The second area is the manpower structure, which consists of 10 variables related to degree, age, or class. The third area is the project nature, which consists of 7 variables closely related to project size. 6) To estimate relationship between productivity measures and measurable characteristic variables of five military R&D centers, and to find regression equations for each productivity measure. With three classes of partial factor productivity measures and seven output factors is 3 and 7, we find 21 regression equations. 7) To demonstrate the usefulness of these models by an example. Benefits of these models include the following: (i) Considering three types of input factors and static as well as dynamic productivity index, the generality of the model`s application does exist. (ii) Since all the input data have been accumulated as tangible and countable, the acquisition and evaluation of pertinent is quick and easy. (iii) Since weights of output factors based on R&D team leaders` opinion and characteristics of respective organizational unit already reflected in the model, the acceptability of evaluation (results) tends to be relatively high.