Applications Of Soft Computing For Sustainable Manufacturing Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Sustainable manufacturing being a trend in current industrialization. It involves of material management with minimum consumption of resources, reducing the environmental impacts from product design and development, and producing product that fulfilling the demands of consumer. In short, sustainable manufacturing emphasized on three main elements; social, economy and environmental. Meanwhile, applying soft-computing approaches into sustainable manufacturing being an important manner because of its capability in solving sustainable related issues have been shown in recent journal articles, review paper, and special issues. However, the applications of soft-computing approaches still being inconclusive and the direction of applying it still not well defined especially in sustainable manufacturing field. Therefore, this article is aims to provide a comprehensive review and investigate the applications of soft-computing approach based on a sustainable manufacturing 6R model. Total of 156 articles have been surveyed in engineering fields to identify the details of soft computing application in sustainable manufacturing. Although this review may not exhausive previewing all the collected data, but it is a valuable information for researchers or academias who are interested to explore in depth in this field of study.

Keywords: Application; Soft-Computing Approach; Sustainable Manufacturing

1. Introduction

In past decade, the concept of sustainable or sustainability was shown as a popular phrase used in various fields or sectors [1][2][3]. People were getting interested to use the 'sustainable' or 'sustainability' as a term to descript product, process, event or concept in different fields of studies that related to social, economy and environmental aspects [3][4].Sustainability has started being defined by the Brundtland Commission as' development that meets the needs of the present without compromising the ability of future generations to meet their own needs' [5]. While in the academic filed, research related to sustainable or sustainability area was first started during 1994 with the works on general view of sustainability in the context of global environment and development needs [6]. From there, the growth of sustainable was extended to different disciplines and fields. While in manufacturing sector, 'sustainable manufacturing' being a hot term to descript the concerns of manufacturer towards green thinking [7].

Sustainable manufacturing brings a meaning of producing product in green production environment, economically, and minimum harm to health [7]. However, implementing sustainable manufacturing into the existing industry sectors being a challenge because of it will deals with various problems from multiple complex manufacturing systems and environment [8][3][9]. Sometimes the implementation of new manufacturing system or concept might fail in the end of its implementation due to some uncontrollable or unknown reasons [10]. On the other hand, implementing new manufacturing philosophy needs a lot of compromisations such as redesign the manufacturing system, understanding the approaches needed to be implemented in its supply chain system [7]. Although there is no single solution that able to answer the concept of sustainable manufacturing, but there are different approaches work in the similar philosophies such as green manufacturing and cleaner production [reference].Recent efforts on Environmentally Conscious Manufacturing and Product Recovery (ECMPR) do showed the concerns on manufacturing activities with the environmental impacts and solutions [11][12] .But, to success implementing sustainable manufacturing, alternative tools or support approaches is a vital element to aid towards sustainable manufacturing. For instance; applying soft computing approach into sustainable manufacturing.

In general, soft computing approaches have been applied to different kinds of applications from science and technology, engineering design, medical, industrial robot and etc. However, applying these knowledge-based approaches into sustainable manufacturing is still in an unclear status [10] .According to the researches being done, the difficulties of implementing sustainable philosophies into manufacturing elements mostly are come from the lack of understanding the concept of sustainable and issue regarding know-how to integrate these philosophies into the manufacturing system [1][13]. Although there are researches being carried out in solving environmental issue using artificial intelligent (AI), but the problem of integrating soft computing still existing and less being discussed by researchers [14]. Therefore, this paper is aims to review and provide an insight of current research trend of sustainable manufacturing with relate to soft computing application.

2.0 Research Methodology

Basically, this research has been carried out through structured methodologies.

Firstly, setting the research boundaries becoming a vital part in the initial stage. This research has been carried out based on specific boundaries in order to drive the research in right route. Indeed, several of the guides have been set by authors to look into deep of the review:

I. The literature search will only base on specific terms search in keywords, abstract, title or topic.

II. Data collection was mainly based on journal article as the sources of this review paper.

III. The databases used for data collection are Web of Science, Scopus, Science Direct, Springer Link, and Emerald Intelligent.

IV. The review is only focuses on the research trend from year 2000-2010.

V. The field or area of research is only covered engineering related disciplines.

In order to include and strengthen the filter during article searching, the keywords have been specified into several groups as shown in Table 1.

Group Keywords Database Results

From 5 databases Related to Sustainable Manufacturing

1 'Fuzzy Logic' AND

'Reduce' OR 'Reducing'

'Reuse' OR 'Reusing'

'Recycle' OR 'Recycling'

'Recover' OR 'Recovering'

'Redesign' OR 'Redesigning'

'Remanufacture' OR 'Remanufacturing' WoSc 53,28


2 'Neural Network' AND

'Reduce' OR 'Reducing'

'Reuse' OR 'Reusing'

'Recycle' OR 'Recycling'

'Recover' OR 'Recovering'

'Redesign' OR 'Redesigning'

'Remanufacture' OR 'Remanufacturing' 154,76

3 'Genetic Algorithm' AND

'Reduce' OR 'Reducing'

'Reuse' OR 'Reusing'

'Recycle' OR 'Recycling'

'Recover' OR 'Recovering'

'Redesign' OR 'Redesigning'

'Remanufacture' OR 'Remanufacturing' 160,

4 'Particle Swarm Optimization' AND

'Reduce' OR 'Reducing'

'Reuse' OR 'Reusing'

'Recycle' OR 'Recycling'

'Recover' OR 'Recovering'

'Redesign' OR 'Redesigning'

'Remanufacture' OR 'Remanufacturing' 15,

5 'Artificial Immune System' AND

'Reduce' OR 'Reducing'

'Reuse' OR 'Reusing'

'Recycle' OR 'Recycling'

'Recover' OR 'Recovering'

'Redesign' OR 'Redesigning'

'Remanufacture' OR 'Remanufacturing' 1

This review paper is aims to collect and analyze only related sources from journal articles because of the data quality and accuracy in describing the information. These articles have been reviewed by international experts based on the latest approaches, methodologies and evidences from various case studies to test it reliability and validity. From there, these data will be analyzed according to the flow in Figure 1.

Figure 1: Methodology Flow-Chart of Review Paper Development

2.1 Terminology

Gaining the basic understanding of the terms architecture in this review paper being important to get a clear picture of the philosophies behind these terms. In this stage, clarifications of the specific terms are needed especially to provide an overview of these terms. Different sources have been gathered to strengthen the term definitions especially from the journal articles. As in this review paper, sustainable manufacturing and application of soft computing are the two main terms needed to identify its conceptual meaning in manufacturing engineering field. These specific concepts need to be classified accurately due to its definitions are vary in different fields and to avoid misleading of the knowledge.

2.1.1 Sustainable Manufacturing: An Overview

Nowadays, producing product in a sustainable ways becomes a must in most of the enterprises due to the challenges of resources scarcity, competitiveness from the global competitor, strict environmental legislations and demanding from the stakeholders [15]. In the chain of sustainable manufacturing, three major elements are focused such as product, process and also system [7]. To start designs a product, designer being the most important person in deciding a product functionality, features, performance and aesthetics look. The influences of designer can lead towards a product successfully be marketed or fail to the market. In fact, more than 80 percent of the products are designed with a short life cycle and contributing to the environmental issues due to the discard processes and product disposal [reference]. While approximate 90 percent of these products are designed to use in one time [reference].

During the process stage, traditional or conventional manufacturing process consumes larger energy and resources during in the cycle chain. This concept has contributed to a huge amount of wastage and giving a deep impacts to the environmental.

However, from most of the previous researches, case studies or even the review papers that related to 'sustainable' in engineering field did mentioned that future trend of the research in sustainable related field especially deal with science and technology or engineering probably will having a close relationship or needs of integration with other approaches such as soft computing approach [1][16][17][12][18][19][20] .

Previously, soft-computing is applied to solve complex problems, predicting optimum solutions and etc in a particular aspects or specific problem such as optimizing a particular parameter in process or design [reference] . But, due the growth of sustainable manufacturing, the parameters dependants are getting huge and much difficult to control and because manufacturer needs to have a deep understanding and know how to get their enterprise sustainable in term of economy, social and environment aspect. However, soft-computing or expert system has shown its capability in solving multiple problems and it already become a future trend of integration in any field of study including 'sustainable manufacturing'.

Sustainability or sustainable is the term that has been discusses and used widely in the field of engineering, economic and business, and also social science related aspect. There is no specific meaning of sustainability since its definition is varying from different disciplinary [1]. However, the focus of sustainable related fields are still in a low concerns and less researches have been done in this specific research area such as sustainable manufacturing [reference]. One of reason is due to the complexity of the sustainability development either to fulfill the criteria of policy or the methodologies of taking into action in the real working environment [1].

The specifications of sustainable manufacturing based on 6R's concept

Figure: Evolution of Manufacturing System

Source: Sustainable manufacturing: Modeling and optimization challenges at the product, process and system levels [7]

Form the trend of research in 'sustainable manufacturing', it is clearly shows that the research articles in this field still very demanding and less. Figure below indicates that total of 19 articles are fulfilling the specific search criteria.

(a) (b)

Figure (a): The distribution of published research in past ten years; (b): Citations rate of the article in 2000-2010

Source: Web of Science Database

Keyword search: 'Sustainable Manufacturing'

Document Type: Article and Review

Subject Areas: Engineering, Manufacturing or Engineering, Industrial or Engineering, Environmental or Engineering, Mechanical or Engineering.

From the initial search, it is clearly shown very less research being done in this specific area from year 2001 to 2006. But, the number of research starts to increase slowly from 2007 to 2009 and the number of article increased double to 4 journals in 2010. Concurrently, the citations rate also shown slow rate from 2003 to 2009 with less than 10 citations rate. While the rate numbers in 2010 is increased to 30 citations. It indicates that future citations rate will be double the number in year 2010.

2.2 Business Competitiveness and Sustainable Manufacturing

In business field, previous trends are working towards returns on investment (ROI) and profit basis. But to remain competitive, they are complying with strict legislation on green elements, consumer preferences and competitor's pressures [9][21]. In a manufacturing system, design stage has been classified as the most important and influence phase to decide whether a product is designed in an economical way, cost affordable, and environmental friendly [9][22]. Although implementing sustainability elements into the an existing business is not a mandatory in certain industries or enterprises, but turning a traditional business into a sustainable business or green business will add credits to the investor especially in term of their competitiveness to the global players [23].

The main barriers or challenges that currently occurred or blocking the involved parties to re-think on how to modify and adopt the elements sustainability into their business with minimum investment [24]. Without clearly understand the concept of sustainability in business, investor sometimes are misleading of the concept of sustainability in business. For these reasons, some investors are rather to maintain their existing business philosophies rather than taking risk to explore something that might make them troublesome. Joanne et al. summarized five challenges that contribute to the lacking of implementing sustainability concept into an enterprise or business [24]:

i. Engage meaningful participation with community members and other stakeholders

ii. Develop beneficial synergies across levels of scale

iii. Produce plans with integrated and sustainable outcomes

iv. Go beyond 'end-of-pipe' to eco-systemic solutions

v. Develop capacity in stakeholders and practitioners to meet the above challenges

From these challenges, it is clearly showing nowadays business philosophy is sticking on the legislations or policies and global trends especially from the advanced countries. Green business which is focused on environmental and economic being a well establishes philosophy and considered as one of the global trends [25].

Besides, product service system (PSS) is also aims to serve a green related business. It is related to sell a service rather than selling a product. The benefits from PSS actually in perfectly control the flow of waste from product's parts by limiting the units own by single user, increasing the value of used part to reproduce as remanufacturing part, preventing or reducing the rate of impacts contributed to the environment along a product life cycle such as pollution due to product improper use. By providing these services in PSS, users and service providers having a close relationship especially in term of understanding the life cycle of a product from the time it has been used until it reached their predicted scarp stage [26]. From there, it creates a tradeoff between environmental and product usage.

2.3 Innovation and Sustainable Manufacturing

In brief, innovation can lead towards producing creative product that able to fulfill the element of sustainable manufacturing (social, economic,enviroment). It can't be denied that innovation is one of the support units in sustainable manufacturing.

Product innovation is one of the key success for most of the enterprise nowadays. As we know, most successful organizations such as Dell, Ford, Toyota, Toshiba and

other multi-national companies are rapidly in figure out their new ideas of innovation to cope the market demands and add-values to their product so that they are able to keep survive in the global market from household appliances to commercial use products. From these successfulorganizations, they are having a similar characteristic such as 'flexible integration' capabilitywhich enables them moving towards the opportunities of tender-based competition. For them, seeking innovative may help the organization to work upon the competitive base and always being the leaders of business [27]. Although some of the new product maybe having same functions as previous model, but minor innovations still considered as the initiative of an organization to compete with other competitors towards advancement of technology.

Basically, the need for new product development & innovation is a must and continuous process in this global market to upgrade product's performance as shown in S-Graph (Figure 2) .

Figure: S-Curve of New Product Development


Product innovation is not just a term of creating something in a creative way, but it is extended to solve problem in an enterprise with limited resources and provide value added information in creating a sustainable manufacturing system.

Although product innovation sometimes is conflicting with the environmental policies but recently it seems most of the enterprise are linking innovation with their product especially for marketing and promoting their concern towards environmental purposes through eco-innovation product such as low consumption energy product.

Figure: closed-loop production system

Source: Synthesis Report [28]

2.3 Background of Soft Computing Approach

The concept of soft computing started was started in early of 1990's by Dr. Zadeh with initially focus on three main approaches in soft computing; Fuzzy Logic; Genetic Algorithm; Artificial Neural Network [reference].

Soft computing approaches or sometimes called as expert systems, artificial intelligent or knowledge based system. It has been recognized as a system that consists of algorithms, heuristic, decision rules and other computation coding. The application of soft-computing is applied in vary fields such as engineering, science & technology, medical or event environmental study. The initial uses of soft-computing approaches are to accelerate the process, analysis, or simplifying the complex work [29]. However, the capacity or the ability of soft-computing is keep growing and potentially to integrate in different field of studies such as social science in dealing with the qualitative or quantitative data analysis [reference].

Generally, there are 3 mains intelligent approaches have been used in engineering field. From most of the previous researches, fuzzy logic, artificial neural network, and genetic algorithm being the most common approaches in soft computing [29][30]. To summarize the differentiations and use of each approach, author has summarized these data into Table 2.

Table: Soft Computing Approach and Function

No Soft Computing Approach Function

1 Fuzzy concept Decision making for incomplete/uncertainties

2 Neural Network Based on historical approach to make adjustment/train to better solutions

3 Genetic Algorithm Case basic, react according to characteristic/objective functions have been set

Source:[31] Integration of fuzzy logic , genetic algorithms and neural networks in collaborative parametric design

But, there are other computing approaches such as particle swarm optimization and artificial immune system have been implemented and developed to solve particular problems or complex issues. Saridakis and Dentsoras [31]simplified the applications of soft computing into two categories:

a. Search for Optimal of Solution

b. Search for Simplification of Solution

Search for Optimal of Solution

Searching the optimal solutions can be difficult due to the uncertainties or hidden factors that might not been considered or

The optimal solution can be

Search for Simplification of Solution

Solving complex problems especially dealing with various parameters or variables that

When dealing with huge/complicated data. Managing these data will be troublesome and time consuming.

There is abundant of literatures mentioning about the soft computing approach in engineering applications. However, the results shows on 67 articles and review are collected based on a specific searching specification. Figure below shows that the publication rate are still low with more than 5 articles published annually along the period but there are only 2 articles published on year 2002 and 2004. While the trend of citations/year was increasing dramatically from less than 10 in year 2002 until more tha 100 citations in 2010.

(a) (b)

Figure (a): The distribution of published research from 2000-2011; (b): Citations rate of the article in 2000-2011

Source: Web of Science database

Keyword search: 'Soft Computing'

Document Type: Article and Review

Subject Areas: Engineering, Manufacturing or Engineering, Industrial or Engineering, Environmental or Engineering, Mechanical or Engineering.

2.3.1 Application of Soft Computing in Environmental and Sustainable related Fields

Basically, the application of soft-computing in environmental and sustainable related fields can be divided into four main categories [14];

i. process control

ii. prediction or estimation

iii. process modeling or simulation

iv. process management

In particular, technologies of expert systems, fuzzy logic, and neural networks have emerged as the most frequently employed approaches for realizing process control [14] .Most of the soft computing approaches are emphasized on the process control [14] . In industry waste reduction purpose, Jeffries et al. [32] have proposed fuzzy logic approach in the control system and investigated that adjustment from the control system is much easier than redesign the existing product layout.

3.0 Sustainable Manufacturing 6R Model

Jawahir et. al. [7] emphasized that traditional 3R concept that has been implemented in 1990's shown a lack in deliver a clear massage especially in an enterprise supply chain. Therefore, they have revised and reconstruct the 3Rs model to 6Rs model to resolve the shortcomings.

Generally the 6R concept is needed in the whole product design and development supply chain loop from pre-manufacture, manufacture, use and finally disposal stage as shown in figure below. In general, the all R's are interrelating each and other in different product design and development stages [33].

Figure: Open Loop Supply Chain System with Integration of 6R Model

Source: Sustainable manufacturing: Modeling and optimization challenges at the product, process and system levels[7]

In this model, it emphasizes on six different approaches which are:

i. Redesign

ii. Reduce

iii. Remanufacture

iv. Reuse

v. Recover

vi. Recycle

3.1 Redesign

As usual, the first person who is playing the mail role in product design and development is the designer. In manufacturing field, designer can be classified as product designer, process designer and also system designer. However, their influences on the manufacturing system becoming important because most of the technical decisions are made by them. Although the justifications of good design and bad design are subject to personal preference. For example, a bad product design can't be justified through it ecstatic or appearance but it should extend to it functionality, reliability and the environmental performance during it end of life. Howarth and Hadfield do agree that the roles played by designer are important but their decisions can be influenced by various stakeholders as shown in figure below [34].

Figure: Typical Stakeholders Influencing Designer

Source: A sustainable product design model [34]

Redesign is actually an activity that refers to significance changes of the existing procedure, system or process and etc in order to achieve a new idea or new innovation thinking. In short, redesign process incorporates a series of decision making and knowledge management issues, which involved of various parties [31].

3.2 Reduce

The main ideas of 'Reduce' is fabricating something with minimum or less material, cost, energy, labor and etc. However, reduce nowadays not limits to the pre-manufacturing stage adjustment but it cover the whole system needs such as the reducing the CO2 emission, pollution rate, toxicity extraction, global climate change and etc.

3.3 Remanufacture

Remanufacture is an effective approach to bring back a product to 'look-new' without compromising the raw material and also the original manufacturing process [33]. To establish remanufacturing, it might be different compare to conventional procedures.

Remanufacture and recycle normally happened simultaneously.

3.4 Reuse

In the product life cycle system, part reuse is one of the main features in sustainable manufacturing. Seliger [reference] mentioned that

3.5 Recover

During the recover stage, the activities normally happened is bring back the components or parts for reuse purpose. The process that involved in recover are product disassembly and part dismantle.

On the other hand, part separation and recovery did happened during part dismantle to avoid damaging the part

3.6 Recycle

Normally, this is the last stage of product design and development whereby the disassembled component will be moved for recycle to the pre-manufacture process to gain back the raw material such as extraction of the aluminum material for future product usage. For environmental conscious manufacturing, the basic approach should be related are 3R concept (Reuse, Reduce, and Recycling).

4.0 Results Classification: Application of Soft Computing and Sustainable Manufacturing

The concept of sustainable manufacturing is varying across different disciplines but its main priorities were focused on triple bottom line: environmental, economical and societal elements. However from the collected data, it is clearly shown economic and environmental elements are given more attention compared to the societal element.

4.1 Redesign and Soft Computing Approach

In product development stage, the evaluation of design is subjective to designer preference. There is no any standard to judge the design either it is a good design or bad design. Therefore, product redesign normally happens when there is a need to change the design according to the requests and problem occurred.

Furthermore, design stage being the most effective stage of deciding the product whether is complying with environmental policies/legislations [35].Researchers suggested that product resign or redesign should not emphasized on the design structure but should going depth into its tolerances and other details geometrical aspects [35]. Moreover, to design a robust system approach, physical programming methods can be proposed to aid design preferences into specific weighted objectives [Jami Kovach].

New idea sometimes can be found from the reviewing through existing problem in a product design or system design through fuzzy case based reasoning [36]. In redesign concerns, neural network with be considered as priority to solve quadratic polynomial in non-linear problems [37]. Machining is one of the processes dealing with high volume of waste and energy. Therefore, to redesign or modify machining tools become a critical element to avoid and reduce unnecessary wastages. For that reason, dry machining or Cryogenic machining being developed as the latest machining approach in solving the environmental related issues from machining processes [7]. Neural approximation method was the most familiar

In the chain of sustainable manufacturing, fabrication of tools to support the product, process and system needed to be emphasized to ensure the completeness and reliability of the whole supply chain system. While in concurrent engineering (CE), time is the main criteria to consider especially in the product development stage. Therefore, redesign is a vital part to improve or to upgrade existing product, process or system to a better performance or results.

4.2 Reduce and Soft Computing Approach

One of the aims of sustainable manufacturing is to reduce the resources. Resources can be in the form of energy, material, human capital, cost, machines and etc. However, this is the most effective way as it develops the concept of use based on the needs only. Therefore, wastes generated will be less and the further effort or alternative to solve the waste will be automatically reduced to minimum.

4.3 Remanufacture and Soft Computing Approach

For remanufacture, sometimes it having a close relationship with recovery in some extends [38]. While reverse logistic is the most common field, it is related to tracking the product part to rework or reconditions to near to new product performance with same features. Product disassembly may become difficult if dealing with multiple complex materials integration. Therefore, researchers suggested that deriving the disassembly process into multiple stages with Genetic Algorithm can aid the process to work more efficiently and minimum cost [38] .

4.4 Reuse and Soft Computing Approach

The first wave of neural network started in year 1943 by McCullock and Pitts from the ideas of applying biological neuron as the main concept to solve problem.

4.5 Recover and Soft Computing Approach

Recovery process initially deals with error solving problem such as system, process that not caused by human error [39]. Providing improvement also being a criteria in recovery a product, process and system

Reverse logistic is one of the key term that represented recovering the product or good in its supply chain system [21][38][40]. Product or goods might need to have second inspection or part repair will need to go through the recovery procedures.

4.6 Recycle and Soft Computing Approach

Solving mixed integer linear programming model (MILP) such as recycling in a reverse logistic, GA has shown a great progress and performance. However, recent researches shown integrating multiple of the sustainable manufacturing concept is still lack (only can be done on single optimization/concern) [25][21].

In some of the case, fuzzy system has been integrated to other decision making tool to perform the capability of weighting and evaluating the vectors recycling such as integrating with AHP and CBR approaches to determine the suitability elements in a product for recycling purpose [36]. Researchers in the other field of study still having the same thinking that integration is needed and is very case dependence [reference]. However,

Recycling activity normally is dealing with cost factor. To achieve a tradeoff between cost factor and multiple uncertainties of cost affected factor. For these reasons, researcher [41] and researcher [42] suggested that integrating of robust stochastic and multiobjective programming with decision making approach such as fuzzy.

Integration of fuzzy set theory in qualitative research to evaluate and provide aid in decision making especially dealing with multiple uncertainties and selection of critical criterion [43] [Kun-Lin Hsieh].

Analytic Hierarchy Process and Dephi Method may be is a good attribute to structured and accurately determine the


1 REDUCE (1st Rank in 3Rs) 1. Waste minimisation 1. Minimise resource and energy use during manufacture

2. Fewer materials are used, the less waste is produced

2 REUSE (2nd Rank in 3Rs) 1. Use an item more than once (same function and some functions) 1. It can be from recycling process whereby part can be recycle used reuse help save time, money, energy, and resources without further process

3 RECYCLE (3rd Rank in 3Rs) 1. processing used materials (waste) into new products 1. Reduce the consumption

2. reduce energy usage

3. reduce air pollution

4. modern waste reduction 1.Producing the same raw material from the process

2. involves destruction process

4 REDESIGN 1. To improve the existing features such as less process, materials used 1. Fewer errors

2. Lower costs

3. Better use of existing resources

4. Better patient access to services

5. Improved relationships (both between staff and with patients) because everyone is less frustrated Process Improvement

Lean Thinking

Six Sigma

Organisational Development

Business Process Reengineering


Lean Six Sigma

Layout design

5 RECOVER 1. It is more on product improvements and the treatment to aid the process 1. It focus on maintenance and stabilization 3 mains stages:

earlier/initial recover


sustain phrase

fault detection/dinogsis

6 REMANUFACTURE 1. The process of disassembly and recovery.

2. A series of manufacturing steps acting on an end-of-life part or product in order to return it to like-new or better performance, with warranty to match 1. Bring product back to 'like new'

2. It remains the functions and feature like new product

3. Problems parts are replaced with new part and no repair process. Fault detection

cradle-to-grave analysis (the tracking of materials from their extraction to their return to earth) and detailed knowledge of the composition of the waste

5.1 Descriptive Analysis

Application of Fuzzy Logic in Sustainable Manufacturing

Fuzzy in Reuse

Fuzzy knowledge or fuzzy concept is applied critically in solving the problems related to reduce, reuse and also recycle. While product redesign and remanufacture are considered medium range. Fuzzy has started to be developed according to the set theory with focus on '1' and '0' as the membership function to represent 'yes' or 'no'. From the perspective of product reuse, Fernando et al. [Fernando] mentioned that fuzzy knowledge is suitable to deal with vagueness of product information such as using scorecard to evaluate the problem and objectives to achieve. These uncertainties of information can be caused by the product design being design only fulfilled or bias to certain aspects of sustainability [reference].

Product design for reuse is another concept that supporting the sustainability philosophy through the triple bottom lines elements. This concept should be encouraged and apply to product design stage focusing in reuse considerations [Jianzhi Li]. In conjunction of this concept, Zha et. al [X.F. Zha] suggested that expert design system is needed to support top-down design especially for the assembled products. As usual cases that involved reuse parts is upgrading PC performance using fuzzy as membership function to reduce the cost and part replace that might indirectly contribute to environmental waste [Hwai-En Tseng].

In the case of decision making, integrating of fuzzy with reason base-reasoning is normally been used. Kim et al. [Kim Hua Tana]and Sisi et. al [Sisi Xuanyuan] agreed that fuzzy knowledge is used to help the manager in making decision technology upgrading issue with reusing all the previous information as their guide. Manufacture or manager is preferred to reuse the existing technologies due to the price factor and to sustain their economic. Therefore, to buy a new technology sometimes is not depends to the current trend but it should wider to other aspect.

As usual, fuzzy system alone is unable to deal with multiple constrains because these constrains sometimes are the variables can only be solved effectively through other soft computing approaches such as artificial neural network (ANN) and genetic algorithm (GA). For instance, fuzzy-neural network mode has been applied by Kim et al. [kim hua tana] to support manager to select the optimal investment in technology selection.

For the future trends of research are:

1. Use a more standard representation for fuzziness, with a more solid theoretical basis such as fuzzy concrete domains [Fernando et al.]


In the social science aspect, Delphi method normally is used to a systematic and structured approach to forecast the replies from respondents through communication technique. But, when dealing with huge data that might confusing the researchers, fuzzy Delphi approach has been proposed by Huey and Zhi [Huey-Jiun Wang].

Fuzzy in recycling

Seo and Park [Kwang-Kyu Seo] addressed that product recycling happened during its disposal stage whereby only the selected parts will be proceed for recycling process or remanufacturing. Sometimes, it is difficult to judge the status of product for disposal because of different usage or product's wear and tear. Clustering these information before recycling becomes a must to ease the process of recycling and predict occurred uncertainties such as modified Fuzzy ART neural network and fuzzy inference approach [Kwang-Kyu Seo][ Jianzhi Li].

Yannis et al.[ Yannis A. Phillis] suggested that to assess material recyclability, the main concerns are the data from the expert .Fuzzy inference and rule-based inference are used to measure the recyclability of a product based on the feedback perspectives from designer, consumer or even the experts [][ Jorge MarxGomez].But, The major problem is lack of knowledge of quantitative dependencies [Yannis A. Phillis]. But, Amy et. al [Amy J.C. Trappey] have proved that integrating quantitative and qualitative with fuzzy approach is possible and shown a good results in solving the problem in reverse engineering.

Fuzzy in Reduce

During the manufacturing process, energy and material consumptions are both main issues faced by the manufacturer. Therefore, fuzzy logic approach is introduced to help manufacturers and designers to predict the use of both resources through the rule-based method [H.C.W. Lau]. On the other hand, Gulcin and Orhan [Gulcin Buyukozkan] having the same view on integrating knowledge based approach in new product development (NPD). They stressed that minimization or reduce the resource planning and cost in the initial stage being the most important criterion. As an example in a bottle filling process design, fuzzy knowledge has been implemented to forecast and control of filling process with minimum energy and waste consumption [M. Jeffries]. To reduce task tardiness, fuzzy-inference model is also proposed to reduce the resource allocation time [Kung-Jeng Wang].

Fuzzy membership function is also been used to reduce the cost and part replace during upgrading PC because of the electronic industry is growing very fast and unnecessary replace of these part might indirectly contribute to environmental waste [Hwai-En Tseng].

Fuzzy in Remanufacture

Remanufacturing having close relationship with the recycle process which means it happens during the post manufacture stage of a product. Arindam et al. [aridam roy] claim that a product's performance can be brought back to close original performance if it is go through the remanufacturing process. However, sometimes it is difficult to separate into details to evaluate the part defectives. Fuzzy is used to detect the defectives of product [aridam roy], while Fuzzy cognitive maping is focuses on tracing the product in reverse logistic and improving the accuracy of collection with other device such as RFID [Amy J.C. Trappey]. On the other hand, part remanufacturing is also influenced by its return values and the whole process costs. Therefore, fuzzy reasoning and neuro-fuzzy approach is proposed by Jorge et al. [Jorge MarxGomez] to solve this problem. While integration of fuzzy knowledge with Weibull analysis-time-to-failure is another approach to estimate the product life and the reselling value of product after serviced from defection [M.I. Mazhar].

Fuzzy in recover

Fault detection and recovery is the aspect need to be considered in a manufacturing system. However, to predict the system fault and plan for recovery or maintenance process is troublesome. Manufacturing system faults are unpredictable due it working environment and the person who are handling it. Therefore, Serge and Luis [Serge Guillaume] proposed a fuzzy partitions approach to separate the uncontrollable variables for evaluation to increase the accuracy of the results. Furthermore, the error detection and recovery using neural-fuzzy approach has been designed and tested of it capability to resolve the error by training through the errors [S. Y. NOF].

Fuzzy in redesign

Design problem is extracted through the exploitation of the design knowledge in the context of a space of solution alternatives [K.M. Saridakis and Dentsoras]. Saridakis and Dentsoras are proposed a Collaborative parametric Design with Soft-Computing (CopDESC) to facilitate the collaboration issue among the designers. Quality functions deployment is another concept to serve consumer preference of products/services based on house of quality approach. Fuzzy FMEA based product design system has been developed by Chin et al. [Kwai-Sang Chin] showed that this process is able to reduce the time taken to product design and avoid the iterations issue.

According to Metin, redesign is one of the global trends that consider as flexible approach to solve integrated systems. Therefore, he proposed a combination Fuzzy Axiomatic Design (FAD) and Analytic Network Process (ANP) approach to solve the process management system especially to deal with incomplete information [Metin Celik].

Application of Genetic Algorithm in Sustainable Manufacturing

GA and Reduce

Identifying the resource allocation for manufacturing process, GA is used to select and arrange the sequence of order especially to control and reduce resource usage [Kung-Jeng Wang]. Towards a sustainable manufacturing, the facility layout do contributed as critical factor especially for the energy consumption during material travelling distance. Therefefore, GA is used to identify strategic location to place manufacturing cell and for assembly sub-systems arrangement [Shana Shiang-Fong Smith].

In cost reduction, limiting the supplier numbers and reducing the cost involved related to supplier is one of the economic elements in sustainable manufacturing. Che and Hang [Z.H. Che *, H.S. Wang] have identified genetic algorithm is suitable to deal with multi-criterions parameters, for instance; product with various parameter requirements or product with special part requirements. Furthermore, genetic optimized decision making is applied to simplify the complicated control of production (COP) process to resolve the product customization and the special part orders issues [W.K. Wong].

Exergy emissions reduction is a complex problem because of the involved parameter and variables are vary and uncertain. Therefore, to solve this problem, Wang et al. [Weimin Wang] have proposed multiple-objective GA with integration of life cycle methodology approach to aid in monitoring the exergy consumption. Genetic algorithm based on mean value has been used to analyze the relationship of cost and energy reduction and the direct relationship has been identified [Jiangfeng Wang].

Product disassembly is a process of dismantle complete product into sub-parts or sub-systems according for recycling, reuse or remanufacture purposes. Review paper focusing on product disassembly done by J.Dong and Graunt [reference] claims that future trends of product disassembly should focused on the integration of soft-computing approach especially to disassemble complex systems with the considerations of cost'..

Lee et al. introduced a hybrid genetic-algorithm to aid the multi-stage disassembly process and reduce the cost needed.

GA and Reuse

Optimal reuse of existing system or product is

GA and Recycle

GA and Redesign

It is becomes a difficult task to find better design that able to compromise sustainability criteria. Redesign the existing product, process or process can becomes one of the alternatives.

Optimal design can lead to minimization of cost and better performance. Genetic algorithm is one of the solutions providing the optimal results.

Apostolos [Apostolos Papanikolaou] and [ reference ] have used the same approaches to solve the redesign in engineering fields with integrate of multiple objective approach.

Conceptual design is to generate the best design with consider of combinatorial optimizations from different parameter.Bo et al. [Ruifeng Bo] have proposed a combinational approach to aid the designer in exploring possible solutions through applying genetic algorithm concept with combination of ant colony optimization and the traveling salesman problem (TSP). Results showed that combinatorial explosion problems in conceptual design can be solved in more operable ways.

In virtual engineering whereby all the works are applied virtually to test its functionality or to monitor the occurring defects before proceed to the real process. Neural-fuzzy method was used for rule-based classification and feature selection in redesign add-on feature [J. Paetz]. The trend of virtual engineering has become popular as it can reduce the errors and time needed to invent a new process or system through real time virtual modeling. Neural-fuzzy is an intelligent methodology finding virtual screening rules in the beginning to reduce the possible uncertainties.

GA and Recover

Product maintenance and recover stand the same meaning to prevent and maintain the performance of a product or system. In robot industry, genetic programming is normally can be seen to generate robust error recovery to maintain its accuracy and reliability [Cem M. Baydar].

Lee and Chan [C.K.M. Lee] mentioned that recycling and recover sometimes happened concurrently especially in the reverse engineering because genetic algorithm is able to propose good and suitable locations for these product to collect and process.

For instance applying multi-objective genetic algorithm to identify multiple better tradeoffs in manufacturing process [Weimin Wang].

Rough set theory is a novel data mining approach that deals with vagueness and can be used to find hidden patterns in data sets [C.L. Huang].

Conclusion and Recommendation

6.1 Limitations

From the review being done, it was found that several limitations have been drawn or determined:

The term being use to search is very limited to the field of study. It should be widen to all related sub key terms of sustainable manufacturing such as product life cycle production.

For soft computing approach, the limitations are the applications are very wide and the difficulties to allocate all these data.


Integration of soft-computing or expert systems into manufacturing system are getting important and being a must due to the complexities of involved uncontrollable parameters and uncertainties especially in the area of sustainability (e.g., economy, social, environmental) [14]. From the empirical studies, it found that not much researches being carried out directly to explain about the integration of soft computing approaches into sustainable manufacturing [reference].

Although the application of soft computing approaches can be easily found the research field especially regarding to engineering related solutions. However, the application of soft computing approaches in sustainable manufacturing are limited or not been showed in a clear manner. It can be concluded that sustainable manufacturing still having a great potentials gaps to explore whereby the numbers of researches in applying soft computing in sustainable manufacturing still very rare and haven't promoted directly. The future work might be more towards developing an integration model or framework to enable an effective communication between soft computing approach and sustainable manufacturing [reference, reference, reference].