The Energy & Resources sector is being reorganized into a trend of strengthening integrated energy system technology centered on low carbon and security, including hydrogen, renewable energy, core minerals, and resource circulation, as well as app...
The Energy & Resources sector is being reorganized into a trend of strengthening integrated energy system technology centered on low carbon and security, including hydrogen, renewable energy, core minerals, and resource circulation, as well as applying digital conversion technology to non-traditional oil and gas development. Considering structural changes in energy industry and shift in software-driven paradigm, support for convergence Research and Development (R&D) in Energy & Resources and Information and Communication Technology & Software (ICT & SW) sectors is essential. R&D in two sectors has a large-scale investment and mid to long term period characteristics, but quantitative performance analysis studies have not been sufficiently accumulated. Multidimensional performance evaluation research should be preceded in preparation for policy decisions related to the allocation of resources such as budgets and regulations in the strategic technology sector, along with enhancing efficiency in consideration of investment in new R&D programs.
This study attempted to conduct an efficiency analysis of national R&D programs encompassing Energy & Resources and ICT & SW sectors by using Data Envelopment Analysis (DEA) suitable for performance evaluation centered on multiple output indicators of R&D in convergent strategic technology. Six DEA models were designed based on technology classification and R&D programs. And then, the relative efficiency of each Decision Making Unit (DMU) was analyzed and the slope between variables, inefficiency factors, and improvement target (%) of the higher efficiency group were derived. As a part of an auxiliary analysis, the relationship between two indicators was analyzed by representing Scale Efficiency (SE) and productivity, the number of papers reflecting weights per 100 million won, by plotting it as a quadrant distribution.
As a result of the analysis, first, in terms of the average efficiency of each model, Model 5 (standard classification of science technology), Model 4 (source R&D program), and in terms of DMU share with efficiency 1 (100%), Model 5 was derived as a major benchmark target, and the efficiency value and reference sets for each detailed DMU were analyzed. Second, for the group with a higher efficiency of 30% or more based on the R&D program model in Banker, Charnes, Cooper (BCC), slope value of the number of papers and patents compared to research budget was estimated at 1.15, 0.58 and the number of papers and patents compared to R&D programs was 5.29, 2.74.
As a result of the inefficiency analysis, first, when considering the inefficiency factors of Pure Technological Efficiency (PTE) and Scale Efficiency (SE), Model 1 (Energy & Resources and ICT & SW technology classification), 2 (climate energy R&D program), 5 and 6 (futuristic promising technology classification(6T)) were found to be a mixture of inefficiency factors, Model 3 (ICT R&D program) was found to factor of SE, and Model 4 was found to be most SE as the main cause of inefficiency. Second, when considering the status of Returns to Scale (RTS), Model 1, 3, 4, 5 and 6 were found to have a dominant share of Decreasing Returns to Scale (DRS), and Model 2 was found to have an Increasing Returns to Scale (IRS). Third, about 42.8% of the top 30% efficiency groups in each model confirmed that the improvement target (%) due to excessive input was not large within the range of -50% to 0, and the improvement target (%) due to underproduction was not large within the range of 0 to 100%. In addition, as a result of productivity quadrant analysis, Model 4 can be considered to be benchmark, and Model 2 can be considered to be avoiding benchmark.
The analysis methodology of this study is expected to be used as a basis data for verifying the efficiency of major performance indicators for each R&D program, as well as the priority allocation of research budget and R&D programs suitable for the technical sector and major classification configuration when planning new technology-converged R&D program and designing policy in two sectors.