Conference Dates

June 5-10, 2016


Researchers have been working on the application of life-cycle assessment (LCA) to solid waste management systems for over two decades. Over this time, the state-of-the-art of LCA has advanced considerably, yet major challenges remain in the use of LCA to evaluate actual solid waste systems. The objective of this presentation is to present some perspective on the major accomplishments to date, applicable modeling techniques and the challenges that remain. This presentation is intended to serve as a catalyst for informal discussion in small groups throughout the conference.

There are many potential uses for LCA models with a considerable range in the required complexity. Perhaps the simplest application is their use as an educational tool to teach high school students and university undergraduates about solid waste and the need to consider environmental impacts. More detailed LCA models are required to effectively evaluate the environmental impacts of alternate policies. For example, in the U.S., the EPA adopted a waste hierarchy in the 1990s that is not supported by LCA analyses. LCA may also be used to guide a city or region in the evaluation of alternatives for solid waste management. Over the past decade, the NC State research group has had the opportunity to work with the State of Delaware as well as Wake County, NC, U.S. on solid waste planning. Such real case studies have served to identify challenges with the application of life-cycle models to real systems. Finally, from a research perspective, models may be used for scholarly pursuits that may include methodological research to improve the overall application of LCA to waste management.

The use of LCA models in all of the applications described above has helped to identify challenges and limitations. One challenge is the tradeoff between simple models that have a limited number of model inputs and more complex and flexible models that require relatively large amounts of input data. While simple models are user friendly and accessible to less experienced LCA and solid waste practitioners, they may oversimplify to the point where the results are not reliable or they may not be flexible enough to consider variations in processes or input values. Nonetheless, simple models can quickly provide first-order comparisons, and may serve to initiate new users in life-cycle thinking. While more complicated models overcome these limitations, they typically require more time and effort to learn to apply properly.

Additional challenges for applying life-cycle models to SWM include data uncertainty, particularly as it applies to the benefits of using recycled materials as a raw material. There is tremendous uncertainty in the available upstream data. Beyond the numerical uncertainty, there are issues related to the location of a process which influences its environmental impact. For example, the emissions associated with plastic or fiber manufacture may be very different in different countries with different emissions regulations. The location of emissions may also significantly alter the risks to human health and the environment. Consider the case of aluminum, which is perhaps the most valuable material to recycle based on energy savings. The energy savings likely occurs at mines and smelters that are distant from the point of use, while additional emissions may be associated with the extra vehicle to collect recyclable materials. Similarly, the benefits of recovered energy from waste may be distant from the solid waste facility at which energy is generated.

Another challenge is tradeoffs between different emissions and impacts. While much work has focused on greenhouse gas emissions, there are examples in which a solid waste management alternative that minimizes greenhouse gas emissions does not minimize, for example, eutrophication or toxicity potential. Additional challenges include the complexity of LCA results and the need to convey simple summary information, and the fact that waste composition, the energy grid, and environmental policies are all likely to change over the relevant time horizon. For example, there is a strong policy focus on landfill diversion of waste, but if paper in the waste stream continues to decrease and food waste continues to increase, then SWM strategies will need to adapt to even maintain current diversion rates.

While there are not perfect solutions to the challenges identified above, the use of sensitivity analyses including contribution analysis, parametric analyses, and Monte Carlo t echniques can help to assess the key inputs and assumptions and robustness of results. The use of life-cycle optimization models has also helped to develop and evaluate novel SWM strategies that minimize environmental impacts or economic costs, while meeting user defined constraints (e.g., diversion targets, budget or emission limits). The application of modeling-to-generate alternatives (MGA) in these models facilitates exploration of the variability in alternative strategies, if any, that exist to meet the same goals. Multi-stage optimization models have also been developed that that allow multiple factors (e.g., waste generation and composition and fuel and electricity mix and prices) to change with time.

Included in

Engineering Commons