Automatically weighted high-resolution mapping of multi-criteria decision analysis for sustainable manufacturing systems

A common feature of Multi-Criteria Decision Analysis (MCDA) to evaluate sustainable manufacturing is the participation (to various extents) of Decision Makers (DMs) or experts (e.g. to define the importance, or “weight”, of each criterion). This is an undesirable requirement that can be time consuming and complex, but it can also lead to disagreement between multiple DMs. Another drawback of typical MCDA methods is the limited scope of weight sensitivity analyses that are usually performed for one criterion at the time or on an arbitrary basis, struggling to show the “big picture” of the decision making space that can be complex in many real-world cases.This work removes all the mentioned shortcomings implementing automatic weighting through an ordinal combinatorial ranking of criteria objectively set by four pre-defined weight distributions. Such solution provides the DM not only with a fast, rational and systematic method, but also with a broader and more accurate insight into the decision making space considered. Additionally, the entropy of information in the criteria can be used to adjust the weights and emphasise the differences between potentially close alternatives.The proposed methodology is derived generalising a problem of material selection of automotive parts in metal casting manufacturing systems. In particular, three typical aluminium, magnesium and zinc alloys in a High-Pressure Die Casting (HPDC) process are compared using the deterministic Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) combining 18 criteria organised in 4 main categories (cost, quality, time and environmental sustainability). A detailed and systematic approach to calculate the considered criteria is also provided and it includes Life Cycle Assessment (LCA) considerations. Results show that, although in most of the cases the aluminium alloy is the best option, there are a few areas in the decision making space where magnesium and zinc alloys score better without a simple correlation to categories. This shows how valuable the proposed mapping process is to understand the complex MCDA analyses. The methodology does not make specific assumptions about metal casting and can be applied to sustainable manufacturing in general.