Steelcase, Inc., a U.S. based and globally operating furniture company, has a long history of environmental improvement throughout its processes and products. Because its products are the core source of these impacts, integrating environmental metrics into the product development process has become a critical effort at the company.
Evaluating the environmental impacts of products can be challenging. Products are typically evaluated through a life cycle analysis (LCA) after design is complete. While this analysis is critical for public reporting and informing future products, a product cannot be revisited to improve performance once it is ready for production. Instead, evaluation of impacts needs to be an integral part of the product development process when materials, processes, and design options can be selected based in part on their expected environmental performance.
This research looked at the feasibility of using a data-driven environmental analysis tool, with the working title of Wizard for Environmental Life Cycle Evaluation (WELE), to reduce the time required for environmental decision making during product development and to minimize the uncertainty of evaluation results when a product design is incomplete. Based on discussions with Steelcase representatives, a beta version of the tool was created within an existing LCA software package and tested with Steelcase product developers to determine its usability. Additional research explored the integration of Steelcase-specific evaluation methods and product data needed to increase the tool's accuracy in reporting environmental impacts. Several iterations of the tool were developed and tested with Steelcase representatives in Grand Rapids, Michigan and Strasbourg, France as well as IDEO, an affiliated product design consulting firm. Separate product tests were also conducted using completed LCAs for existing Steelcase products. These tests included evaluation of the impacts on full product performance when generic versus company-specific materials and processes were used. They also included modeling of the products in increasing detail to determine potential levels of reporting accuracy at each stage of product development.
This research indicated that there is value in using a data-driven approach to environmental analysis in early stage product development, but there are also several challenges. The product tests demonstrated that representative estimates of environmental impacts can be achieved in the early stages of product development, even when multiple design decisions remain to be made. Across the tests, environmental impacts represented at each stage of product development were compared with the products' final LCA results. In the concept phase of development, 18 (or 32% with a modified product) - 63% of final impacts were represented. This moved up to 50 - 80% of impacts represented in the design phase, 62 - 92% represented in the engineering phase, and 95 - 99% represented in the final production phase. While these results were promising, several challenges also emerged regarding the tool's usability as well as long term data collection and management. Therefore, while the data-driven approach has many benefits, improvements to the non-expert usability of LCA platforms and development of data collection efforts will be essential to optimize such an approach.