Measuring incineration plants’ performance using combined data envelopment analysis, goal programming and mixed integer linear programming

Abstract

Incineration plants produce heat and power from waste, reduce waste disposal to landfills, and discharge harmful emissions and bottom ash. The objective of the incineration plant is to maximize desirable outputs (heat and power) and minimize undesirable outputs (emissions and bottom ash). Therefore, studying the overall impact of incineration plants in a region so as to maximize the benefits and minimize the environmental impact is significant. Majority of prior works focus on plant specific decision making issues including performance analysis. This study proposes a hybrid data envelopment analysis (DEA), goal programming (GP) and mixed integer linear programming (MILP) model to assess the performance of incineration plants, in a specific region, to enhance overall power production, consumption of waste and reduction of emissions. This model not only helps the plant operators to evaluate the effectiveness of incineration but also facilitates the policy makers to plan for overall waste management of the region through decision-making on adding and closing plants on the basis of their efficiency. Majority of prior studies on incineration plants emphasize on how to improve their performance on heat and power production and neglect the waste management aspects. Additionally, optimizing benefits and minimizing negative outputs through fixing targets in order to make decision on shutting down the suboptimal plants has not been modeled in prior research. This research combines both the aspects and addresses the overall performance enhancement of incineration plants within a region from both policy makers and plant operators' perspectives. The proposed combined DEA, GP and MILP model enables to optimize incineration plants performance within a region by deriving efficiency of each plant and identifying plants to close down on the basis of their performance. The B Prasanta Kumar Dey

Role of artificial intelligence in operations environment: a review and bibliometric analysis

“…It acts like a human to solve analytical problems, mimics visual and speech issues, understands and removes language difficulties. Four main elements of expert systems cover knowledge base (explains the collection of facts, rules as desired from human experts), inference engine (talks about how to solve problems critically), scheduler (takes care of serials of a problem in question) and user interface (enables adequate interaction between various programs that run in parallel) (Petridis and Dey, 2018; Shibin et al , 2018; Singgih et al , 2018; Sheng et al , 2019). Further, components of genetic algorithms highlight its importance (Stefanini et al , 2018; Tseng et al , 2018).…”

Flue Gas Emissions and Performance Evaluation of Small-scale Solid Waste Incinerators at Njokerio and Ng’ondu in Njoro, Kenya

“…This section includes findings of solid wastes incineration at varying: moisture contents; incinerator waste loading rates; and operating temperature levels on flue gas emissions. The incineration of solid wastes was influenced by the sizes, shape of combustion chambers, wall insulations, turbulence levels, loading rates and methods of air injections as reported by Petridis and Dey [4]. The low waste heating values, low temperature levels and lowered turbulence resulted into rising up of residence/holding time to complete combustion process and increased flue gas emissions similar to what Astrup et al [1] reported.…”

“…In their study, Petridis and Dey [4] reported that the incineration performance was subject to wastes quantity and quality which varies over time in certain locations due to variability in solid wastes generation and seasonality.…”

Multi-Level Data Envelopment Approach for Performance Evaluation of MSW Generated in Urban Area of Madhya Pradesh, India

“…The efficient frontier is built utilising linear programming methods using observed input and output data. Data Envelopment Analysis (DEA) is the most extensively used non-parametric approach [19,[24][25][26][27].…”