Most manufacturing enterprise should perform environmental impact assessment throughout the entire life cycle of their products due to the increased environmental consciousness of customers and the introduction of strict environmental regulations. Life cycle assessment (LCA) is a systematic method of analyzing the environmental load of a product and evaluating potential environmental effects. It can be useful for eco design and for dealing with environmental regulations. However, the LCA process generally requires considerable time and money to collect relevant life cycle inventory (LCI) data and information. Usually, enterprises develop a new product by revising or reusing a similar previous product. Repeating the LCI data collection whenever they develop a new product is very cumbersome and will increase lead time. However if we estimate the eco impact values based on the previous design, then the efforts related to LCI data collection and its attending complex computational procedures are reduced. Although various efforts have been made on the streamlined LCA in an effort to overcome these limitations, the result is still unsuitable for practical eco design. We therefore propose an approximate LCA method using case-based reasoning (CBR) for a rapid and convenient environmental evaluation in product development. For the approximate LCA, we developed function behavior structure environmental effect (FBSE) representations and a creative similarity measurement based on FBSE for a clear and consistent CBR process. A geometry attribute based linear modeling algorithm of eco impact is proposed to replace complicated LCA procedures, and genetic algorithms are used to search for optimal solutions to satisfy the proposed model. A case study involving an upstream process of a vehicle air purifier fan confirms that the proposed CBR for LCA method can be effectively applied to eco product design.