A niched genetic algorithm to solve a pollutant emission reduction problem in the manufacturing industry: A case study

doi:10.1016/j.cor.2005.08.005

Computers & Operations Research

Volume 34, Issue 7, July 2007, Pages 2191-2214

https://doi.org/10.1016/j.cor.2005.08.005 Get rights and content

Abstract

A multiobjective optimization approach to deal with a pollutant emission reduction problem in the manufacturing industry, through implementation of the best available technical options, is presented in this paper. More specifically, attention is focused on the industrial painting of wood and the problem under investigation is formulated as a bicriteria combinatorial optimization problem. A niched Pareto genetic algorithm based approach is used to determine sets of methods, tools and technologies, applicable both in the design and in the production phase, allowing to simultaneously minimize the total cost and maximize the total pollutant emission reduction.

Introduction

In the last few decades, environmental problems have received much attention from society and governments. In this context, the European Union has introduced the concept of $B$ est $A$ vailable $T$ echnologies ( $BAT$ , for short) as a new integrated approach to deal with environmental problems caused by pollution produced by manufacturing/production industries.

This approach highlights the requirement that the development of new manufacturing/production processes must be balanced by the implementation of best available technical options, that eliminate or at least reduce the amount of pollutants released into the environment.

The $BAT s$ are all the methods, tools and technologies that can be applied in the design and production phase, in order to achieve a good reduction of emission level with an acceptable additional cost.

Often, different types of technical options can control a given source. In this context, mathematical programming techniques can be used to select the best combination of removal and control technologies from different alternatives available.

Depending on the specific goal to be achieved, different optimization problems can be formulated [1], [2]. Generally, the main objective is to minimize the total cost of policy decisions (i.e., mainly the costs involved in implementation and maintenance of removal and control technologies), whereas, environmental considerations are addressed through integrating emission constraints. Other modelling approaches handle the minimization of the emission as another objective function in addition to the cost. The resulting multiobjective optimization problem is then converted to a single objective problem by a linear combination of the different objectives as a weighted sum.

In this paper, we consider a multiobjective combinatorial linear programming model that handles pollution emissions and control costs simultaneously as competing objectives. The main aim is to determine the Pareto set of combinations of $BAT s$ that allow both to maximize emission reductions and to minimize the total control cost.

For the solution of the problem under investigation, a niched Pareto genetic algorithm based approach is defined and implemented. The effectiveness of the developed algorithm is evaluated by using an illustrative case study, that is, industrial painting of wood.

The remainder of the paper is organized as follows. Section 2 gives the general mathematical formulation of the problem we are dealing with. Section 3 presents some basic concepts used in multiobjective evolutionary approaches and gives a general description of the niched Pareto genetic algorithm. Section 4 describes the considered manufacturing process. Section 5 introduces the niched Pareto genetic algorithm specifically designed for the industrial painting of wood. Section 6 reports the experimental results obtained on a real case study. Finally, Section 7 presents the conclusions of this work.

Section snippets

Problem statement

In order to describe the problem under investigation, we focus our attention on generic manufacturing plants that can be viewed as decomposed in different production phases, consuming some inputs in order to produce some outputs.

The scheme of a generic manufacturing plan is outlined in Fig. 1, where the production phases are modelled as black boxes, whereas the raw materials represent the inputs for the entire process.

During the execution of successive production phases, additional materials

Evolutionary approaches to multiobjective optimization

Evolutionary algorithms are search procedures that imitate the process of natural evolution in order to solve complex optimization problems. Since the 1960s, several evolutionary methodologies have been proposed that can be broadly grouped in: genetic algorithms [7], [8], [9] evolutionary programming [10] and evolution strategies [11], [12].

The main idea of the mentioned approaches is to process a set of candidate solutions, called population, simultaneously. Population is modified by two basic

The manufacturing process

We have focused our attention on the industrial painting of wood. Our choice has been motivated by the fact that the directive 1999/13/CE of the European Commission [31], [32] applies to the companies working in this productive sector. Such a directive focuses on the emissions of volatile organic compounds (i.e., ${VOC}_{s}$ ) into the environment by manufacturing industries and, in order to prevent or reduce the direct and indirect effects of such emissions, the directive has established limit values

Design of NPGA for industrial painting of wood

The first step in designing a genetic algorithm for a particular problem is to devise a suitable scheme to represent the individuals. In the $NPGA$ , proposed for solving the problem under investigation, the set of technical options available for each phase is viewed as an individual or member of the population. Consequently, a binary representation is an obvious choice for the problem since it represents the underlying $0 - 1$ integer variables. Hence, in our representation, we use a $n$ -bit binary

Numerical results and discussion

In this section, we present the results obtained from the application of the $NPGA$ on the problem described in Section 4, by considering as case study the manufacturing process of an Italian company specialized in the industrial painting of wood.

The main activity of the company in question is the manufacturing of chairs. In this case, the threshold value for the consumption of solvents, established by the 1999/13/CE norm, is fixed at 15 tons per year. The consumptions are determined by

Concluding remarks

In this paper, a $NPGA$ based approach has been applied to the pollutant emission reduction problem in the manufacturing industry. The problem under investigation has been formulated as a bicriteria optimization problem with competing cost and environmental impact objectives. The sensitivity of the proposed solution approach to the parameters that control the behaviour of the algorithm itself have been assessed, namely population size, tournament size and niche radius.

The computational results,

References (35)

H.I. Shaban et al.
An optimization model for air pollution control decision making
Environmental modelling and software with environment data news
(1997)
W.W. Cooper et al.
Survey of mathematical programming models in air pollution management
European Journal of Operational Research
(1996)
Roth B, German BJ, Mavris DN, Macsotai N. Adaptive selection of engine technology solution sets from a large...
C.A. Coello Coello
A comprehensive survey of evolutionary-based multiobjective optimization techniques
Knowledge and Information Systems
(1999)
C.A. Coello Coello et al.
Evolutionary algorithms for solving multi-objective problems
(2002)
C.A. Coello Coello
Evolutionary multiobjective optimization: current and future challenges
D.E. Goldberg
Genetic algorithms in search, optimisation and machine learning
(1989)
Davis L, editor. Handbook of genetic algorithms. New York: Van Nostrand Reinhold;...
J.H. Holland
Adaptation in natural and artificial systems
(1992)
L. Fogel et al.
Artificial intelligence through simulated evolution
(1966)

I. Rechenberg

Evolutionsstrategie: Optimierung technischer Systeme nach Prinzipien der biologischen Evolution

(1973)

H.P. Schwefel

Numerical optimization of computer models

(1981)

K. Deb

Multi-objective optimization using evolutionary algorithms

(2001)

Schaffer JD. Some experiments in machine learning using vector evaluated genetic algorithm. PhD dissertation (TCGA file...

J.D. Schaffer

Multiple objective optimization with vector evaluated genetic algorithms

Fourman MP. Compaction of Symbolic Layout Using Genetic Algorithms. Proceedings of the first international conference...

C.M. Fonseca et al.

Genetic algorithms for multiobjective optimization: formulation, discussion and generalization

Cited by (13)

A holistic review of applied methodologies for assessing and selecting the optimal technological alternative from a sustainability perspective
2014, Journal of Cleaner Production
Assessing and selecting the optimal technological alternatives in industrial sector is a fundamental tool to improve and adapt the industrial processes to the European legislation, which encourages the incorporation of new technologies for minimising the environmental impacts of companies such as Best Available Techniques. The aim of this study is to perform a holistic analysis of this process from a sustainability perspective. That is to say, its purpose is to analyse the studies that have been published in the literature over the last decade concerning the way the different improvements incorporated into different industrial sectors (Best Available Techniques, green technologies, remediation technologies, emission abatement technologies, etc.) are assessed and selected from a range of perspectives (environmental, technical, economic, social, etc.). Thus, an extensive review of the literature was performed to determine which criteria are the most widely used, the methodologies applied to choose those criteria, the sources of information the data is obtained from, the methodologies for calculating the indicators that allow each criterion to be measured, the different normalising and weighting methods that are applied, the techniques that are most commonly used to compare and choose the best technological alternatives and, finally, the sensitivity analyses that are most often applied to such processes. As a result, the process for assessing and selecting technologies is characterised and the guidelines to direct the selection of the set of methodologies that are the most appropriate to make a decision related to technology selection in future applications are established.
Optimization of global and local pollution control in electricity production from coal burning
2012, Applied Energy
Citation Excerpt :
Ren et al. [12] introduced a multi-objective approach based on linear programming (LP) for the design of a distributed energy system that minimizes simultaneously an economic objective function (the annual energy cost) and an environmental one (the annual CO2 emissions). Grandinetti et al. [13] presented a multi-objective linear programming model to identify the optimal set of Best Available Technologies (BATs) in an industrial facility for painting wood that accounts for the simultaneous minimization of the emissions and total cost. In the same line, Mavrotas et al. [14], proposed a methodology that offers the decision-maker combinations of BATs to be implemented.
World electricity generation is expected to strongly increase over the next 25 years, mainly due to the demand of the emerging economies. In this context, coal is likely to remain as one of the most important energy sources. Several pollution control devices have been proposed so far as Best Available Techniques (BATs) to retrofit coal combustion plants in order to minimize their environmental impact. In this work we have developed a systematic method based on multi-objective optimization for the optimal design of pollution control devices in coal-fired plants that considers simultaneously economic and environmental criteria. Our formulation includes nonlinear models for predicting the cost and performance of the abatement technologies. The design task is formulated in mathematical terms as a bi-criteria mixed-integer nonlinear programming (MINLP) problem, the solution of which is defined by a set of Pareto points that represent the optimal trade-off between the two objective functions. The applicability of this method is explored through a case study based on a pulverized coal power plant of 500 MW (gross), burning low-sulfur bituminous coal. Numerical results show that it is possible to reduce the impact of these facilities at a marginal increase in cost by adjusting the operating conditions and capacities of the control pollution devices.
Optimization of heat pipe with axial "Ω"-shaped micro grooves based on a niched Pareto genetic algorithm (NPGA)
2009, Applied Thermal Engineering
A niched Pareto genetic algorithm based approach is utilized to optimize a heat pipe with axial “Ω”-shaped micro grooves. The effects of the structural parameters are evaluated and optimized with respect to the heat transfer performance in order to model the heat transfer capability and total thermal resistance of this novel heat pipe. Using the heat transfer capability and total thermal resistance as the objective function and the structure parameters as the decision variable, the optimization design for the heat pipe is performed using the niche genetic algorithm. The results indicate that the heat transfer capability and the total heat resistance are inversely coupled and as a result, the optimization must be constructed on the application objective. Using the niched Pareto genetic algorithm and the pre-specified application constraints, a Pareto-optimal solution set can be determined and the optimal design for a given application is selected.
A dynamic chain-like agent genetic algorithm for global numerical optimization and feature selection
2009, Neurocomputing
Citation Excerpt :
The algorithms for global optimization problems are very important in modern engineering design and system operation in many areas. Among the existing optimization algorithms, the genetic algorithm (GA) has received considerable attention regarding its potential as a novel optimization technique for complex problems and has been applied in various areas successfully [2,4,5,7,9,11]. The main specific feature of the GA as an optimization method is its implicit parallelism, which is a result of the evolution and the hereditary-like process.
In this paper, one novel genetic algorithm dynamic chain-like agent genetic algorithm (CAGA) is proposed for solving global numerical optimization problem and feature selection problem. The CAGA combines the chain-like agent structure with dynamic neighboring genetic operators to get higher optimization capability. An agent in chain-like agent structure represents a candidate solution to the optimization problem. Any agent interacts with neighboring agents to evolve. With dynamic neighboring genetic operators, they compete and cooperate with their neighbors, and can use knowledge to increase energies. Global numerical optimization problem and feature selection problem are the most important problems for evolutionary algorithm, especially for genetic algorithm. Hence, the experiments of global numerical optimization and feature selection are necessary to verify the performance of genetic algorithms. Corresponding experiments have been done and show that CAGA is suitable for real coding and binary coding optimization problems, and has more precise and more stable optimization results.
Toward a robust optimal point selection: a multiple-criteria decision-making process applied to multi-objective optimization using response surface methodology
2021, Engineering with Computers
An integrated optimization and simulation approach for air pollution control under uncertainty in open-pit metal mine
2019, Frontiers of Environmental Science and Engineering

View all citing articles on Scopus

View full text

Short communicationA niched genetic algorithm to solve a pollutant emission reduction problem in the manufacturing industry: A case study

Abstract

Introduction

Section snippets

Problem statement

Evolutionary approaches to multiobjective optimization

The manufacturing process

Design of NPGA for industrial painting of wood

Numerical results and discussion

Concluding remarks

Environmental modelling and software with environment data news

Survey of mathematical programming models in air pollution management

European Journal of Operational Research

A comprehensive survey of evolutionary-based multiobjective optimization techniques

Knowledge and Information Systems

Evolutionary algorithms for solving multi-objective problems

Evolutionary multiobjective optimization: current and future challenges

Genetic algorithms in search, optimisation and machine learning

Adaptation in natural and artificial systems

Artificial intelligence through simulated evolution

Evolutionsstrategie: Optimierung technischer Systeme nach Prinzipien der biologischen Evolution

Numerical optimization of computer models

Multi-objective optimization using evolutionary algorithms

Multiple objective optimization with vector evaluated genetic algorithms

Genetic algorithms for multiobjective optimization: formulation, discussion and generalization

Short communication
A niched genetic algorithm to solve a pollutant emission reduction problem in the manufacturing industry: A case study