Product Design Support System

Product design focuses on individual customers

In recent years, customer needs have become increasingly varied. Very often, people will not favor a product that is owned by many others. It follows that companies need to develop a range of products whose designs differ in ways that meet various user needs.

This research area aims to develop a system that investigates customer needs and helps developers or designers to design affective products.

The development of computer simulation and CG technology has made the process of product design more efficient and effective. On the other hand, product designers must now consider not only performance or function but various other elements such as appearance, feel of the material, impression and so on.

Consider, for example, the design of a car.

The original function of a car was to carry people and baggage and to move in safety. On that basis, the key features of a car were driving performance, power performance, space design and safety design.

But while exterior and interior design do not influence a car’s performance, these are also very important design factors because some people view a car as luxury goods. In that context, it becomes very important for customers that, for instance, the sound of a door opening and closing is satisfying.

And because the quality of that sound is influenced by the design of both door and body, it changes with the design parameters for door and body. But because there are so many design parameters, it is very difficult for designers to identify those that will optimize the sound of a door opening and closing.

Computer simulation and parameter optimization algorithms offer an effective solution to this problem. A system that simulates door sounds on the basis of design parameters can optimize those parameters to develop a pleasing sound.

In this system, the computer randomly generates some initial sets of design parameters and then uses simulation to create sounds from those parameters. By listening to those sounds, a designer can evaluate them to identify which are “good” or “bad.” Based on those evaluation results, the computer gradually optimizes the design parameters, repeating this process to arrive at the most pleasing sound.

Anyone can be a designer

We have referred only briefly to design support systems in this car design example, but design work is becoming common for all of us. Many people now have a website to display their content. To make it as attractive as possible, anyone creating a web presence must consider multiple design components such as logos, fonts, layout, colors and coordination.

Our system can automatically design web pages by combining these various design components. The only user task is to evaluate the designs suggested by the system.

The system also creates texture designs for web page backgrounds. This includes design parameters for texture such as roundness, squareness, colors, positions and so on, automatically generating texture designs from those parameters. In this way, the system helps users to arrive at a more attractive design.

Study themes

We are exploring a basic technology for interactive evolutionary computation and developing some applied systems.

Studies include a tournament-style evaluation interface for multiple users’ votes, a recommendation system using Kansei retrieval agents and a cooperative design support system.


T.Urai, M.Tokuamaru, “User Kansei Clothing Image Retrieval System”, Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol.18, No.6 pp. 1044-1052, 2014-11.

H.Takenouchi, M.Tokumaru, N.Muranaka, “Interactive Evolutionary Computation Using a Tabu Search Algorithm”, IEICE TRANSACTIONS on Information and Systems, Vol.E96-D, No.3, pp.673-680, 2013-03.

H.Takenouchi, M.Tokumaru, N.Muranaka, “Tournament Evaluation System Considering Multiple People’s Kansei Evaluation”, Journal of Kansei Engineering International, Vol.9, No.2, pp.43-50, 2010-06.

H.Takenouchi, M.Tokumaru, N.Muranaka, “Tournament Evaluation System Applying Win-Lose Result Presumption Considering Kansei Evaluation by Multiple People”, Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol.16, No.3, pp.453-461, 2012-05.

H.Takenouchi, M.Tokumaru, N.Muranaka, “Performance Evaluation of Interactive Evolutionary Computation with Tournament-Style Evaluation”, WCCI 2012 IEEE World Congress on Computational Intelligence, pp.193-200, 2012-06 (Brisbane, Australia).

H.Takenouchi, T.Hirokata, M.Tokumaru, N.Muranaka, “Running Shoe Design System with Interactive Evolutionary Computation”, International Conference on Kansei Engineering and Emotion Research 2012, pp.925-932, 2012-05 (Penghu, Taiwan).

Y.Inagawa, J.Hakamata, M.Tokumaru, “A framework of recommender system considering the variety of Kansei”, Joint 6th International Conference on Soft Computing and Intelligent Systems and 13th International Symposium on Advanced Intelligent Systems (SCIS&ISIS2012), pp.197-202, 2012-11 (Kobe, Japan).

D.Okunaka, M.Tokumaru, “Kansei retrieval model using a Neural Network”, 12th International Symposium on Advanced Intelligent Systems – ISIS2011, pp.483-485, 2011-10 (Suwon, Korea).