The University of Birmingham
School of Computer Science

Industrial Project from Systems Biologica for MSc in Natural Computation (2009/10)


Multi-objective product carbon footprint optimisation

As pressure from consumers and legislators continues to increase, many organisations are beginning to see assessment of their carbon footprint as vital to their future success. Understandably, the focus to date has been on the carbon footprint of an organisations own operations. However, the majority of greenhouse gas emissions typically fall outside of a single organisation's boundaries and occur up and down the supply chain. Product carbon footprinting is thus gaining increasing attention, which allows the calculation of the carbon footprint of individual products or services taking into account their full life cycle from raw material through to final disposal.

Systems Biologica's ecompete software service allows an organisation to calculate product carbon footprints and identify emission "hot spots" within the products life cycle, which are a natural focus for emission reduction strategies. In some cases reducing emissions within these hot spots requires consideration of a number of interacting variables either within an individual process and/or the wider supply network i.e. emission reduction often requires solving a complex optimisation problem. Additionally, reduction of emissions must be achieved while controlling associated costs.

This project will involve consideration of one such multi-objective optimisation problem and will consist of two phases:

1) Mini project; Literature review of carbon footprint optimisation relating to product life cycles. This may involve individual processes within a product life cycle and/or their supply networks. Identification of a suitable optimisation problem to focus effort in phase 2. This will likely be a problem that has been studied previously with known variables and existing published results.

2) Full project; Utilisation of suitable optimisation algorithms, e.g. multi-objective genetic algorithms, to optimise the chosen problem with respect to both emission reduction and cost. Production of a range of solutions with differing trade-offs between these two variables. Conclusions relating to general applicability of this approach.


Page maintained by Xin Yao.