Toward Quantitative Evaluation of Public R&D Investments: A Framework Concerning Japan’s Semiconductor Strategy

Abstract

Semiconductors serve as a cornerstone of national industrial infrastructure and are critically important strategic technologies and materials from the perspective of economic security. In recent years, particularly with regard to next-generation semiconductors, governments around the world have provided unprecedented levels of support, intensifying international competition. However, public investment decisions in the technology sector may rely heavily on internal heuristics within government institutions, and resource allocation is not always optimal. As the physical limits of the miniaturization approach, the importance of back-end processes is increasing, and evaluation metrics are becoming more diverse. In the absence of systematic use of indicators such as Figures of Merits(FOMs) and Technology Readiness Levels (TRLs), evaluation criteria vary across projects, leading to inconsistencies in resource allocation priorities and potentially reducing the overall efficiency of public support. This study adopts a systems thinking approach to compare semiconductor policies across countries and focuses on technology roadmapping as a method for enabling quantitative evaluation using FOMs and TRLs. Based on this methodology, the study proposes an evaluation framework for government Research and Development (R&D) projects. Specifically, it analyzes the status of FOM implementation within Japan’s nextgeneration semiconductor policy and examines the feasibility of quantitative decisionmaking through a hypothetical case study focusing on high-bandwidth memory (HBM). The comparative analysis of Japan and the United States reveals that Japan’s top-down, concentrated investment strategy offers strengths in terms of speed and consistency, while the US model excels in long-term infrastructure development, broad stakeholder engagement, and flexible resource allocation. The study also finds that Japan lacks a comprehensive national strategy that clearly outlines investment priorities by technology area. As a result, resource allocation tends to be fragmented and dispersed. Furthermore, at the policy level, technology roadmaps based on FOMs and clear strategic guidance are insufficient. At the project level, many initiatives lack quantitative goals and evaluation metrics, and the use of FOMs remains limited and inconsistent. To address this issue, evaluation, and portfolio management methods that incorporate FOMs and TRLs may provide a basis for improving alignment with policy goals and enabling transparent and rational investment decisions. This study suggests that developing national-level technology roadmaps and establishing a consistent evaluation system based on FOMs represent promising strategies for enhancing the strategic coherence and operational efficiency of Japan’s semiconductor policy.