Page 222 - Materials Chemistry, Second Edition

220 10. Advancing life cycle sustainability assessment using multiple criteria decision making

method seems to be a more appropriate technique due to its simplicity and pictorial
representation of relationships.
Step 6: Consistency of inputs on indicators: Partial check for preferential interdependence
must be checked. In case there exists dependency within the indicators, indicators
selection can be revisited in Step 1. For example, combine multiple indicators with
dependence or break dependent indicators into multiple subindicators. Similarly,
inputs must be checked for asymmetry, transitivity, and comparability, and if
negative values of indicators exist, then normalization of data is appropriate.
Step 7: Selection of weighting methods for indicators: Different weighting method produce
different results. Therefore, use of multiple weighting methods is recommended
and carrying out sensitivity on weights.
Step 8: Apply MADM method using the weights for indicators and impacts (Scores) from
Step 3 to obtain integrated final impact.
Step 9: Uncertainty and sensitivity analysis: As there are number of sources of uncertainty
in LCSA based MADM approach, detailed uncertainty and sensitivity analysis
should be performed.
Step 10: Interpretation of results from the analysis: MADM results are usually obtained as
single scores and hence need further interpretation. This is an essential step where
the final scores should be validated with the data and methods used for LCSA.
Step 11: Report the final aggregated impact with recommendations and share the results
with stakeholders for feedback.

10.7 Conclusions

LCSA is a fast-developing field, and numerous efforts are being made to refine the frame-
work and associated methods used for sustainability assessment. In this work, we have taken
stock of using different MADM methods for LCSA. The basic structure of LCSA is described
in detail and highlighted the suitability of MADM methods in integrating indicators
with LCSA.
The review of applications of MADM for LCSA showed that there are numerous chal-
lenges of applying MADM to LCSA. The challenges of MADM application are discussed
in detail. A framework is proposed for carrying out LCSA using MADM. The framework
is also able to highlight tackling of challenges in integrating MCDA with LCSA, such as, dom-
inating alternatives, choice of appropriate MADM method, consistency of inputs on indica-
tors, selection of weighting methods for indicators, and uncertainty and sensitivity analysis.
One of the critical issues identified is the choice of MADM method for LCSA. It is
recommended that there is no unique suitable MADM method for LCSA, and hence,
it is suggested to define scenarios for the given decision-making situation in LCSA. Once
the scenarios are articulated, accordingly, more refined weights can be given to the indicators.
Using this set of weights, if more than one MADM method ranks the same alternative as most
preferred then such an alternative can be conclusively identified as more sustainable than the
other one based on LCSA coupled with MADM approach. In addition, there exist many meth-
odological uncertainties while implementing LCSA (choice of assessment methods, data