Volume13
Abstract: Data analytics and machine learning have the potential to modify and improve many old school businesses. Among the oldest businesses for the human race is managing the waste we generate. In this paper we show how data science can be applied to help derive increased value from a byproduct of that waste, landfill gas. Gas produced from the decomposition of waste in landfills can be captured and transformed into a resource that benefits the local community, environment, and economy. We use analytics to better understand how weather conditions impact the methane content of landfill gas in ways significant enough to interfere with its use as a source of energy. We model methane concentrations in landfill gas and use machine learning techniques to predict future changes in methane concentration using a database of weather, water composition, and landfill gas collection performance metrics. A multilayer predictive model of methane concentration is developed that will aid in the transformation of day-to-day operations of landfill gas collection to maximize the utilization of gas extracted from the landfill, while minimizing the cost of pollution mitigation. This can help transform the industry while mitigating some environmental concerns. Download this article: JISAR - V13 N3 Page 29.pdf Recommended Citation: Hassler, E., Emery, D., Hoyle, J., Cazier, J., (2020). Trash to Treasure: Predicting Landfill Gas Flow to Optimize Electricity Generation. Journal of Information Systems Applied Research13(3) pp 29-39. http://JISAR.org/2020-3/ ISSN : 1946 - 1836. A preliminary version appears in The Proceedings of CONISAR 2019 | ||||||