Usage Context-Based Design (UCBD) is an area of growing interest within the design community. Usage context is the set of scenarios in which a product (or service) is to be used, including the environments in which the product is used, the types of tasks the product performs, and the conditions under which the product is purchased and operates. It is proposed in this work that in choice modeling for usage context-based design, usage context should be a part of the primary descriptors in the definition of a customer profile, in addition to the socio-demographic attributes for modeling customers' heterogeneity. As customers become more technology-savvy and market-educated, current choice modeling methods in engineering design could greatly benefit from exploiting the rich contextual information existing in product usage. In this work, we propose a choice modeling framework for Usage Context-based Design (UCBD) to quantify the impact of usage context on customer choices. We start with defining a taxonomy for UCBD. By explicitly modeling usage context's influence on both product performances and customer preferences, a step-by-step choice modeling procedure is proposed to support UCBD. Two case studies, a jigsaw example with stated preference data and a hybrid electric vehicle example with revealed preference data, demonstrate the needs and benefits of incorporating usage context in choice modeling.