Unleashing Analytics to Reduce Costs and Improve Quality in Wastewater Treatment

Alexander Zadorojniy (), Segev Wasserkrug (), Sergey Zeltyn () and Vladimir Lipets ()
Additional contact information
Alexander Zadorojniy: IBM Research, 3498825 Haifa, Israel
Segev Wasserkrug: IBM Research, 3498825 Haifa, Israel
Sergey Zeltyn: IBM Research, 3498825 Haifa, Israel
Vladimir Lipets: IBM Research, 3498825 Haifa, Israel

Interfaces, 2019, vol. 49, issue 4, 262-268

Abstract: Wastewater treatment is carried out in plants using a complex series of biological, physical, and chemical processes. Typically, these plants operate in a conservative and inefficient risk-averse mode that makes it difficult to quantify the risks or truly minimize the costs. We developed an innovative operational control process that applies descriptive, predictive, and prescriptive analytics to improve efficiency and reduce costs. The descriptive analytics use historical sensor data to build a simulation model and plant-state estimator. The predictive analytics model the wastewater treatment process behavior using a transition probability matrix, which we estimated. Finally, our prescriptive Markov decision process analytics offer recommendations for improved operations. We deployed our system at a plant in Lleida, Spain. The results of the pilot showed a dramatic 13.5% reduction in the plant’s electricity consumption, a 14% reduction in the amount of chemicals needed to remove phosphorus from the water, and a 17% reduction in sludge production.

Keywords: Markov decision process (MDP); multivariate adaptive regression splines (MARS); reinforcement learning; wastewater treatment plant (WWTP) (search for similar items in EconPapers)
Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://doi.org/10.1287/inte.2019.0990 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:inm:orinte:v:49:y:2019:i:4:p:262-268

Access Statistics for this article

More articles in Interfaces from INFORMS Contact information at EDIRC.
Bibliographic data for series maintained by Chris Asher ().