The use of circular economy is becoming more and more important, particularly in the field of agriculture, a major provider of waste. In particular, a lot of researches are being done to transform the lignocellulosic waste from agriculture through desired "sustainable" processes. Sustainable processes mean economically viable, socially accepted, and environmentally responsible processes. Thanks to the "life cycle thinking", it is possible to assess such potential environmental impacts. However, these environmental analyzes require a lot of specific data, whose collection can be long and tedious, or simply impossible in practice. On the other hand, the huge amount of scientific articles describing the processes of valorization of co-products of agriculture constitutes a great, largely under-exploited source of data. Knowledge engineering (KE) tools can be used to compile processes and analyze them. In this paper, we propose an innovative approach, based on intensive data and KE methods, to help a decision maker to choose between different pretreatment processes and different biomasses. The main goal is to develop an intensive, semi-automated data collection approach and an associated tool for assistance with choices in a circular economy context. It is defined by five steps: (1) goal and scope, (2) intensive data and knowledge and the allocation of flows and releasesstructuration and integration, (3) life cycle inventory (LCI), (4) sustainability assessment and (5) analysis and ranking. The study of 13 pretreatment processes of rice straw and corn stover validate our proposal.