MODEL DESIGN FOR ANALYZING CUSTOMER PREFERENCE TOWARDS HEELED SHOES DESIGN OF ADORABLE PROJECTS

• Main Author: Mulia, Stella
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/39723
• Institution: Institut Teknologi Bandung

In order to survive in dynamic markets, Adorable Projects launches new product nearly every month. During the new product development process, designers use their intuition in analyzing the customers’ abstract preference. This causes a huge mental load, resulting a design without proper data-driven information. Incompatibility between product design and customer preference will lead to inefficiency of time and resources. Therefore, the aim of this research is to develop a method for customer preference analysis and an application made to facilitate the model execution and results representation. Fuzzy association rule mining is used to discover relationships between product attributes and salability of heeled shoes. Binary particle swarm optimization and genetic algorithm substitute traditional algorithms that require support and confidence threshold. Comparison between the two algorithms shows that genetic algorithm outperforms binary particle swarm optimization with model validation of 80,4%. Afterwards, an application prototype is developed based on the chosen model. This application utilizes Microsoft Excel and R. Verification analysis shows that the application satisfies Adorable Projects’ needs in analyzing customer preference.