My main research field is mathematical optimization. I am particularly interested in problems that require repeated decision making (multi-stage decision processes) and problems involving uncertainty (stochastic models). I have been researching the theory and application of dynamic programming, which is a method to efficiently solve these kinds of problems. I am also engaged in research on social mathematics, which uses mathematical techniques to design fair and highly convincing systems and measures to address social issues. My co-researchers and I have developed technology in collaboration with real world sites of social issues. Two of these case studies are presented below.

The admissions process of matching children to daycare centers takes into consideration not only applicant priority criteria (i.e., each applicant’s childcare needs are calculated as a score.), but also requests for siblings to be admitted to the same daycare center. This is called “Matching with Couples” and is known to be a difficult problem even academically. Figure 1 shows two patterns of applicant dissatisfaction. An excellent seat assignment that does not cause this type of dissatisfaction is called “stable matching.” However even its existence is not guaranteed, and satisfactory adjustments are not easy to achieve.

(a) There is a child with a lower score at my first choice.

(b) Siblings should have been admitted together to their third choice.

Figure 1: Patterns of dissatisfaction among applicants

As a result, each local government required a great deal of manpower and time for trial and error, and in some local governments, problems arose such as an increase in the number of cases where siblings were placed in separate daycare centers. Therefore, Division of Fujitsu Social Mathematics of IMI, together with Fujitsu Laboratories, addressed this issue and proposed a new method (using the theory of extensive form games) to achieve fair seat assignment. The proposed method has been commercialized by Fujitsu Limited and is already in use by 35 local governments as of June 2020．

The logistics industry in Japan is facing a severe shortage of labor, and there is an increasing need for joint transportation allowing large amounts of cargo to be transported using fewer trucks. Therefore, we developed a joint transportation matching technology that instantly lists and proposes combinations of highly efficient combined transportation using a database with many registered transport lanes. For example, round-trip and triangular transportation are effective in reducing the empty backhauls (Figure 2). The lower the empty running rate is, the more efficient the process is. However, browsing through enormous combinations of two transport lanes is time consuming. Therefore, our newly developed technology makes good use of the hidden inequalities, such as the “distance axiom,” to narrow down the search range without sacrificing accuracy. This technology has been installed as a core engine in the joint transportation matching system “TranOpt” provided by Japan Pallet Rental Corporation. As of October 2023, approximately 180 companies are using this system.

Figure 2: Processing a triangular transport matching request