Design for environment as a tool for the development of a sustainable supply chain

Abstract

In this work, design for environment (DfE) methodologies have been used as a tool for the development of a more sustainable supply chain. In particular by combining life‐cycle assessment (LCA) techniques and by using the quality function deployment (QFD) multi‐criteria matrices, an ‘environmental compromise’ can be reached. In this work, the QFD matrices have been developed in a new way using an iterative process that involves the whole supply chain starting from the product life‐cycle, taking into consideration the machines that make the product and their components. This methodology is compatible with the requirements of the various stakeholders, suppliers, manufacturers and clients, involved in the supply chain. To assess the validity of the proposed approach a specific supply chain was studied concerning packaging systems for liquid food substances (beverage cartons). Firstly all the stages which are most critical from the environmental point of view in the supply chain of packaging systems were identified and assessed. The starting point for the analysis of environmental aspects and impacts which characterise the supply chain was LCA, which proved to be useful for the identification and the environmental assessment of the various stages in a packaging system. Through the use of ‘iterative QFD’ it is possible to arrive at a definition of the engineering characteristics of all the machinery which is involved in the supply chain. In particular in this work the authors have tried to identify the critical points in the design of those machines which either make the beverage cartons or are involved in the filling process.

Keywords:

• design for environment
• LCA
• QFD
• packaging systems

1. Introduction

Recently there has been increasing emphasis on the need for industrialised countries to address issues of sustainable development, understood as ‘development which satisfies the needs of the present without compromising the possibility for future generations to satisfy their own needs’. Industries and designers have therefore found themselves faced with the necessity to adopt opportunely new tools and reference parameters for production and design. Design for environment (DfE) is one of the possible approaches. The concept of DfE can be summarised in the principle ‘do more, using less’ which is the incentive for reducing the quantity of energy and material used to provide goods and services. In parallel with the general reduction in this type of consumption there are also problems regarding the use of scarce raw materials and the emission of substances which are harmful for the environment and for man.

Much research focuses on the environmental implications of design decisions and on methods to determine and influence the environmental impacts of products. Less attention is paid to the organisational consequences of such insight (Bakker 2002). Moreover, Nielsen and Wenzel (2002) are of the opinion that some of the most important decisions concerning the environmental properties of a new product are made during product development. Thus, significant environmental improvements can very often be achieved by integrating environmental properties as an optimisation parameter during product development together with parameters such as function, production cost, ergonomics, etc.

Recently, more attention has been paid to environmental supply chain management, defined as ‘the set of supply chain management policies held, actions taken and relationships formed in response to concerns related to the natural environment with regard to the design, acquisition, production, distribution, use, reuse, and disposal of the firm's goods and services' (Zsidisin and Siferd 2001). According to Hagelaar and van der Vorst (2002) life‐cycle assessment (LCA) can be seen as the main instrument of environmental supply chain management; it is a technique for gathering data on environmental care issues, which can be used to restructure supply chains in order to improve their environmental performance. However, these authors conclude that there are no guidelines for this integration.

In the first part of this work, we propose a procedure which, by combining the LCA technique and an appropriate use of quality function deployment (QFD) type multi‐criteria matrices, tries to define design specifications for all the stakeholders involved in a supply chain: clients, manufacturers, suppliers, suppliers of the suppliers, etc.

The second part of this work, using the procedure proposed, studies a specific supply chain for packaging systems for liquid food substances (beverage cartons). Life‐cycle thinking must influence the environmental management of firms which deal with packaging systems involving all areas of their activity, starting from ways of designing the product and finding raw materials, to packaging and machine production operations and even to the management of the post‐consumption/disposal phase. In the case of beverage cartons the filling and packaging of the liquid substances is the central link which unites the various stages in the supply chain for packaging systems for liquid food substances. The design approach proposed in the first part of this work was therefore applied to this type of machinery.

2. Material and methods

Two methods have been used in this work to define the most suitable DfE choices for minimising the environmental impact of a supply chain: QFD, a qualitative method for obtaining the opinion of experts and LCA, a quantitative method for assessing environmental loading.

The QFD matrices used in this study have been adapted from the more general QFD matrix, used as a quality control technique for activities and typical features of a product through a series of subsequent steps (Hochman and O'Connell 1993, Presley et al. 2000).

Life‐cycle assessment methodology was chosen to identify the stages of the supply chain which are the most critical from the point of view of environmental impact. In fact LCA allows an assessment to be performed of the environmental loads connected with a product, by identifying and quantifying the energy and the materials used as well as the waste products released into the environment (Carlson et al. 2001). According to Khan et al. (2002) LCA is one of the most important techniques for the successful implementation of a process or product development in the context of environmental sustainability.

Several authors have used the two above‐mentioned techniques in different ways trying in some cases to integrate them with other techniques, with the aim of creating decision‐making tools. Cristofari et al. (1996) proposed a new methodology using QFD and LCA to document technical requirements. Different alternatives were assessed based on these requirements to select the best product. Akao (1990) described various applications of QFD techniques in product development. Hanssen et al. (1996) and Ferde et al. (1995) applied QFD, LCA and LCC separately for environmentally sound light fittings, but did not form a systematic methodology that could integrate QFD, LCA and LCC into an efficient tool. Graedel and Allenby (1996) developed an environmentally responsible assessment matrix to simplify the LCA process. Costic et al. (1996) estimated the environmental performance of conventional lead‐based solders and their substitutes. Donaldson et al. (1996) conducted LCA for the telecommunications semiconductor laser industry.

Halog et al. (2001) used QFD for the improvement analysis of selected ‘Best Available Techniques’. They developed a modified QFD version and applied it to determine the emissions which need to be analysed further for environmental performance improvement. The target specifications used are the environmental benchmarks obtained from the comparison of emission values of the techniques.

Zhang et al. (1993) proposed a new methodology by integrating LCA, LCC, and QFD into an efficient tool that deploys customer, environmental and cost requirements throughout the entire product development process. Azapagic (1999) also discussed LCA application in process selection and design. On similar lines, Khan et al. (2004) proposed a life‐cycle indexing system – LInX – which facilitates LCA application in process and product evaluation and decision‐making. The LInX comprises four important sub‐indices or attributes – environment, health and safety (EHS), cost, technical feasibility, and socio‐political factors. Furthermore, each attribute contains a number of basic parameters, e.g. EHS consists of 11 parameters. Quantification of each basic parameter is performed for the complete life‐cycle of a proposed process or product. An analytical hierarchy process is used to compute the weights for each basic parameter and sub‐index.

Senthil et al. (2003) developed a life‐cycle environmental cost analysis (LCECA) incorporating costing into LCA practice. This model prescribes a life‐cycle environmental cost model to estimate and correlate the effects of these costs in all the life‐cycle stages of the product. The newly developed categories of eco‐costs are: costs of effluent treatment/control/disposal, environmental management systems, eco‐taxes, rehabilitation, energy and savings of recycling and reuse strategies. The mathematical model of LCECA determines quantitative expressions between the total cost of products and the various eco‐costs.

Huang and Ma (2004) combined three methods to evaluate the impact of packaging materials:

1. life‐cycle assessment (LCA), a quantitative method, to assess environmental loading;

2. analytic hierarchy process (AHP), a qualitative method, to obtain opinions from experts;

3. cluster analysis to integrate the results of the former two methods.

The authors developed this method to provide integrated information and avoid a bias towards either a qualitative or a quantitative approach.

The papers available in literature only take into account a single product rather than all the stakeholders involved in the product life‐cycle. The procedure proposed in this work, using QFD matrices in series, allows the definition of guidelines for the whole supply chain. This is useful in order to provide ‘Engineering Characteristics’, information about the manufacturing cycle and about machinery design for decision makers who are involved in the supply chain.

Each stakeholder in the supply chain (clients, manufacturers, suppliers and suppliers of the suppliers) can reduce the environmental impact of a supply chain by making appropriate DfE choices. The contribution of the various stakeholders involved in the production chain must however be guided in order to define the design choices which will have most effect on the overall environmental impact. This method also tries to arrive at an ‘environmental compromise’ which is compatible with the various requirements of the stakeholders involved in the supply chain: suppliers, manufacturers and clients.

3. DfE procedure in the development of a more sustainable supply chain

The process which goes from the design of a machine to its sale can be divided into five main stages: project definition; concept development; prototype development; field test; and commercial launch.

For each of these stages there are various activities which involve different aspects of design development (technical, legal, economic, environmental, etc.). It is important to insert environmental considerations in all the design development stages, from the moment the original idea is generated up to the time of launching the product on the market.

In this way the targets that a firm must set itself are:

	to guarantee environmental performance of the product for all the stakeholders;
	to ensure respect for environmental law;
	to ensure that its own environmental performance is coherent with the firm's standards and strategy;
	to minimise the environmental impact of the product without compromising other external obligations (cost, safety, functionality);
	to promote the environmental certification of the product.

One critical aspect in machine construction is the understanding of which design choices and parameters exert most influence, from an environmental point of view, over the whole supply chain in which the machines are involved. The aim of this study is to propose a solution by redefining the activities which are carried out at the first stage of the machine manufacturing process: project definition (see Figure 1).

Figure 1 Steps in machine manufacturing design.

The starting input for the project definition stage can be re‐formulated as follows:

	an analysis of legal requirements and internal guidelines for DfE;
	an analysis of the supply chain in which the machine is involved;
	an LCA study of the whole supply chain in order to understand the contribution of each stage in the production chain to the total environmental impact. The output of this step is the definition of the most important recommendations (Whats) for the supply chain.

The following steps concern the development of several matrices the levels of which are increasingly detailed from the point of view of technical specifications (see Figure 2):

	the Whats are transformed into basic engineering characteristics (ECs) of the product in the supply chain. The ECs are the technical answers to the Whats requirements;
	preparation of the relationships matrix (‘iterative QFD’). The team of experts must assess which ECs have an impact on the Whats and to what extent;
	the weights of the ECs are fixed at the bottom of the matrix. They represent the key result of the first matrix and the input of the second matrix: the most important requirements of the product (‘Engineering Characteristics’) in turn will become the recommendations (Whats) which by means of suitable matrices will be implemented for the machines used in manufacturing the product itself;
	the use of a cascade of matrices therefore allows the transformation along the whole supply chain of environmental requirements into ECs which will then be implemented on the machines that are used at the various stages of the supply chain;
	on the basis of the indications provided by the QFD matrices the final stage is that of revising and in some cases of introducing DfE procedures into the machinery
	the output of the second matrix could in turn become the input of a third matrix which would define the ECs of the machine components. Suppliers to machine suppliers are involved in this step.

Figure 2 Development of the multicriteria matrix.

The people involved in the design process of a machine are aware of their role. However, they do not have a thorough understanding of the supply chain as a whole, with a consequent weakness in communication between the parties which makes it impossible to organise environmental improvement plans. To overcome these obstacles and to carry out the project definition activities a panel of experts was formed to encourage communication and meetings where the operators and suppliers could contribute their knowledge and information about the processes.

The engineering choices and characteristics, which result from the ‘Project Definition’ stage, are finally reconsidered and verified in the other stages of machine development.

The ‘concept development’ stage involves:

	assessing to what extent the environmental requirements fixed by the law and by company policy can really be reached;
	assessing the environmental performance of the product using the tools described by ‘corporate standards’;
	verifying whether it is possible to satisfy environmental targets and to conform with corporate standards and restrictions in the use of dangerous substances;
	documenting the results.

The ‘prototype development’ stage involves:

	assessing the impact of any changes made;
	optimising the environmental performance of the product;
	testing and verifying whether the environmental targets and the standards which were considered attainable in the previous stage have been satisfied and whether the risks have been eliminated.

The ‘field test’ stage involves:

	optimising the environmental performance of the final product;
	checking conformity to the laws and to the environmental policy of the firm;
	checking what has been tested and verified during the final steps of the previous stage;
	documenting the results obtained;
	including the environmental information in the technical manuals and sales information.

The ‘commercial launch’ stage:

The aim of the DfE project is to develop the machine using design criteria which respect the environment and to provide clients with information about the environmental performance of the machine. The procedure for certifying that this information has been defined on the basis of some reference standards already present on the (European) market, such as the environmental product declaration (EPD) and ISO 14025 (Allander 2001).

To assess the validity of the proposed approach a specific supply chain has been considered in Section 4: packaging systems for liquid food substances (beverage cartons).

4. Case study

This study focuses on the supply chain for beverage cartons. The beverage carton is the most‐used container in Europe for packed juice and milk. In Europe, some 100 million beverage cartons are filled with liquid (milk, juice, wine, water, soup, etc.) every day (Huat 2005). This type of container protects sensitive products against the effects of oxygen, bacteria and light. Moreover, it reliably protects liquid foods using a minimum of materials.

An LCA study should specify the functions of the system under study. The ‘functional unit’ measures the performance of the functional output of the product system. The main function of the packaging of beverage cartons is to contain the product; while marking the contents and increasing visual appeal are less important. Therefore, ‘g/l’ (grams of packaging materials/per litre of beverage) was selected as the ‘functional unit’ in this case. The beverage densities are all assumed to be equal to 1 g/ml. A 1 litre beverage carton weighs 32 grams. The material used for beverage cartons consists of about 76% wood fibre, 18% polymers and 6% aluminium.

The impact categories used in the study and the ‘weighting’ criteria for the types of impact were determined using the eco‐indicator method (Goedkoop 1995, Goedkoop and Spriensma 1999). Figure 3 shows the tree diagram of the environmental contribution characterising the various processes which a carton undergoes (the case study refers to packaging for fruit juice). The steps involved are summarised in the following sections.

	Aluminium production: aluminium oxide is derived from bauxite extraction and is then transformed into aluminium through an electrolytic process. The aluminium is then laminated and transported to the converting factory.
	Paper production: as is well‐known, the process is based on wood processing.
	Plastic production: the plastic material used is low density polyethylene (LDPE) coming from natural gases and petroleum which, when appropriately treated (refining, cracking, polymerisation, etc.), make up the LDPE flake.
	Beverage carton production: this consists of the assembly of aluminium foil, paper‐board and polyethylene through a process of moulding and roll forming; the rolls are then delivered directly to the clients using over‐land transport.
	The filling process: this takes place directly on the client's premises. The cartons are made up from the rolls of packaging material and then filled.
	Distribution, sale and consumption: the cartons are assembled in cardboard packaging then palletised and delivered directly to the retailer.

For the final stage in the life‐cycle, three possible scenarios have been taken into consideration:

	Recycling: beverage cartons are recyclable. Carton recycling has been growing across Europe since 1992 and now exceeds 250,000 tonnes annually. This equates to 27% of all cartons placed on the market. Through dedicated paper mills, cartons are recycled into other useful products such as tissue and kitchen paper, office stationary, fibre‐board and reel cores for industrial use. The non‐fibrous element of cartons is recovered through waste‐to‐energy plants or extrusion into new products.
	Energy recovery: this derives from the incineration of beverage cartons taken over‐land to specific sites. Cartons have a high energy content, making them very suitable for incineration with energy recovery. In the European Union, in 2000, 25% of all beverage cartons were used in waste‐to‐energy plants to produce steam and electricity.
	Landfill: here the organic remains of the cartons decompose with a partial release of methane gas (belonging to the group of greenhouse effect gases).

Figure 3 Tree diagram showing the environmental contribution of the various processes.

The results of the LCA study are reported as percentages for the impact categories studied [global warming potential (GWP), acidification potential (AP), energy and resource].

Analysis of the results of the LCA study has allowed the relative importance of the various steps in the production line to be estimated, and some useful observations to be made as follows:

	the production of raw materials is responsible for the main contribution in the life‐cycle of the system studied (aluminium, paperboard and plastic);
	the coupled carton production (converting) stage is equally relevant from the environmental point of view, in particular because of the waste coming from packaging material processing;
	during the beverage packaging step energy consumption, principally in the form of electricity, together with the AP parameter, prevail over the other types of environmental impact. Moreover neither the waste of packaging material, which accounts for about 25% of the total environmental impact at this stage, nor the loss of the product (in this case fruit juice) should be underestimated;
	the contribution arising from the product distribution, retail, and consumption stage is mainly caused by the transport and double or triple packaging of the product;
	recycling and energy recovery have a ‘positive’ impact on the environment;
	landfill disposal is relevant as far as GWP is concerned due to the quantity of methane gas released during the decomposition of the packaging material.

4.1 Development of ‘iterative QFD’

The methodology used in this study involved the development of two ‘QFD’ matrices, which by degrees go more and more deeply into the design project and into the engineering processes with increasing levels of specificity.

For this study a panel of 10 participants was formed, which included two academics, whose research studies are mainly focused on environmental management, three technical operators and three managerial operators involved in the design and production processes of the filling machines and two suppliers of filling machine components. The number of participants, which at first sight may seem rather large, derives from the Delphi technique (Linstone and Turoff 1975) adopted for working with panels. The Delphi technique is a structured process which investigates a complex or ill‐defined issue by means of a panel of experts. This methodology proves to be an appropriate research design for this type of research and permits individual opinions to be obtained within a structured group and using a communicative process.

The first matrix (see Figure 4) which was developed on the basis of the LCA recommendations for the supply chain of a packaging system, provides the ECs for the beverage carton product, in terms of their relative importance and the benchmark values that must be reached (Ansari and Modaress 1994, Park and Kim 1998).

In particular the steps followed for drawing up the matrix were:

(1) Definition of the most relevant environmental aspects

When studying the case the group of experts was asked to define some recommendations (‘Whats’) for every step of the supply chain of the beverage carton product. For example, for the first stage of the supply chain, purchasing of raw materials, the principal ‘What’ is: reduce the quantity of the material required to make the product (aluminium, paper and plastic). One or more ‘Whats’ can be defined for each step of the supply chain, see Table 1.

Six evaluation factors (‘Whats’), taken from ISO 14021 (ISO 1999), were considered to assess environmental friendliness.

Table 1. The most important environmental aspects, ‘Whats’.

Supply chain	Customer attributes (Whats)
Purchasing of raw materials	Reduce the quantity of the material required to make the product (aluminium, paper and plastic)
Converting	Reduce the waste of packaging material during the production stages
Filling	Improve the energy efficiency of the processes
Distribution retail and consuption	Optimise the transport and material handling stage
Recycling	Favour the recycling of the used cartons
Energy recovery	Favour the energy recovery of the used cartons

(2) Weighting of the ‘Whats’ by importance

The importance or priority of the ‘Whats’, ‘Imp(What)’, was calculated by the team of experts using the LCA analysis. In particular a sensitivity analysis was developed. Using this analysis it was possible to assess as a percentage the impact on the whole supply chain of a 5% reduction in ‘Whats’. Some rules were also created to convert this percentage value into an importance judgement (see Table 2). The ‘Imp(What)’ was assigned using three levels, ‘low’, ‘medium’ and ‘high’, corresponding to numerical values of 1,3 and 9 respectively.

Table 2. Calculation of the importance of ‘Whats’ (1st matrix).

Rules	Importance of the Whats
If (GWP and AP and energy and resources) decrease by less than 1%	1
If [(GWP or AP or energy or resources) decrease by between 1% and 2%]	3
and
If [(GWP and AP and energy and resources) decrease by less than 2%]
If (GWP or AP or energy or resources) decrease by the same as or more than 2%	9

For example, a 5% reduction in the first What (‘reduce the quantity of material…’) has a relative importance of more than 2% on impact categories along the whole supply chain and it will be associated with an importance value of 9 (see Figure 4). If the total reduction in all the impact category values had been less than 1%, an importance value of 1 would have been associated to this ‘What’, while in all other cases an average importance value of 3 would have been associated. It was hypothesised that the four impact categories all have the same importance.

(3) The ‘Whats’ are transformed into basic ECs

The ECs are the technical answers to the Whats requirements. Table 3 shows the engineering design parameters used for the product. The choice of ECs was made on the basis of interviews with experts as well as on the basis of current laws and the design standards.

Table 3. Product design specifications.

Engineering characteristics (ECs)
• Produce rectangular beverage cartons
• Produce cartons which weigh less
• Replace glass fibre
• Use easily recyclable materials
• Use raw materials coming from renewable resources
• Following design for disassembly methodology
• Produce recloseable beverage cartons
• Replace CFC
• Replace halon

By manufacturing rectangular cartons it is possible to optimise space during the distribution stage.

Plastic materials are used for their light weight, formability and low cost. However, because of their easy degradability over time and in sunlight they need to be improved using various additives. For example, in order to delay their flammability flame retardants are used which are particularly harmful for the environment.

CFC is used in cooling systems and halon in fire extinguishers. Replacing CFC and halon, the substances having the highest ozone depleting potential, with alternatives having no halon depleting potential is very important for environmental policy.

Another important environmental factor is the glass fibre contained in plastic. This prevents recycling because it damages and spoils the mechanical qualities of the reconstituted plastic.

(4) Preparation of the relationships matrix

It is necessary to assess which and to what extent the ECs have an effect on the Whats. The relationships between ECs and Whats, indicated in the ‘cells’ of the matrix may be positive or negative, of weak ( = 1), medium ( = 3) or strong ( = 9) intensity.

(5) Drawing up the correlations matrix

The physical relationships between the specific techniques (ECs) are to be found in the ‘roof’ of the matrix. This step in the construction of the matrix helps in the tracing of pairs of ECs which require parallel improvements and/or those which include ECs in potentially difficult relationships and which therefore involve courses of action which conflict with each other. Therefore this matrix provides positive and negative correlations between pairs of ECs as well as the entity of these correlations.

(6) Other measurements

The product development team must estimate the costs, feasibility and technical difficulty involved in modifying each EC. Objective values must be fixed, that is to say measurements which reflect the connection between Whats and ECs, as well as the clients’ requirements.

(7) Action plan

The weights of the ECs are situated at the base of the matrix. They are the key result of the planning matrix and are calculated using the expression:

where V(EC) _in is the correlation value of EC _i with What_n and Imp(What_n ) represents the importance or the priority of What_n .

The construction of the matrix and the results obtained are illustrated in Figure 4.

Figure 4 1° matrix, level 0.

The results obtained at level 0 are the input of the matrices at detail level 1, refer to Figure 2, that is to say the matrices which link the ECs of the product with the design characteristics of the machines involved throughout the beverage carton production line. This paper does not consider all the machines involved in the production line, but only those which are used during the filling and packaging stages. The choice of the raw materials used in the beverage cartons, the types of distribution, the possibility of recycling and re‐utilising materials are in fact connected with filling and packaging methods. The machines which make beverage cartons and which fill them (filling machines), starting from a roll of pre‐folded coupled material, composed of wood‐fibre, polymers and aluminium have as their output the packaging filled with the product.

Passing from level 0 to level 1, refer once again to Figure 2, the ECs of the first matrix become the ‘Whats’ of the second matrix, see Figure 5. The relative importance of the (ECs) _i has been used to determine the new values of importance for the ‘Whats’ as shown in Table 4.

The values that define the various classes of importance were calculated taking into consideration the ECs number (9 in the case studied) and the number of classes that are required (3 in the case studied). The limit of the first class was calculated as:

while the limit of the second class was calculated as:

Table 4. Importance classes for ‘Whats’ (2nd matrix).

Relative importance ECs	Importance of the ‘Whats’ (2° matrix)
⩽5.5%	1
5.5–16.5%	3
⩾16.5%	9

In order to build the second matrix steps (1) to (7), as previously explained for the formation of the first matrix, were repeated. Table 5 shows the engineering parameters ‘ECs’ for the design of the filling machines used. Even in this case the choice of the EC parameters was arrived at by a team of experts not only on the basis of their knowledge of both present and future legal requirements and design standards which filling machines must satisfy, but above all on the basis of DfE principles.

Table 5. Machine design specifications.

Engineering characteristics (ECs)
• Reduce the volume/weight of the moving parts of the machine
• Minimise the need for heating/cooling during product processing
• Use high pressure nozzles to reduce water consumption by the machine
• Harmless emissions in the air and in water from the production plants
• Design the components of the machine according to design for rotation units
• Choose machine components with the longest possibile lifetime
• Reduce wasting the product (fruit juice) during the filling process
• Replace some solvents

Figure 5 2° matrix, level 1.

5. Discussion

The environmental sustainability of the manufacturing, packaging and distribution process for liquid food substances is influenced by many factors. In recent years, the areas that many firms in the liquid packaging sector have focused on in order to attain important environmental targets are:

	partnerships with suppliers of aluminium in order to produce aseptic packages with less aluminium content but having the same product protection features;
	partnerships with suppliers of plastic materials in order to reduce the emissions deriving from packaging material production processes;
	the introduction of a new package format which is lighter and cheaper, designed for the protection and the aseptic packaging of beverages, in particular of milk, to be delivered in less time to those areas of the world which are the most difficult to reach.

From the analysis carried out in this paper using the first matrix it can be seen that the three most important ECs to implement for beverage cartons are:

1. manufacture lighter beverage cartons so as to reduce the consumption of raw materials;

2. use materials which can be easily recycled;

3. choose design for disassembly methodology.

The drawing up of the matrix of correlations between the various ECs allows us to have an overall view of the integration between the different phases in the life‐cycle of the product and the environmental aspects, thereby ensuring an adequate solution to the trade‐off associated with most decisions which characterise a design. The various EC modification proposals, with their relative importance values, arising from the two matrices must be assessed by estimating from an environmental and economic point of view the effects of the change to a set of product design parameters. Using the roof of the first matrix it is possible to identify the following three types of trade‐off.

	Concerning environmental aspects: minimising the weight of a product sometimes threatens the possibility of recycling. In the last 20 years, for example, improvements to this aspect of the product have resulted in the average weight of a beverage carton being reduced by 20%. Nevertheless, designing for environmental benefit must always be balanced with the need to avoid compromising consumer safety by undermining the true function of the packaging product – its ability to protect and preserve its contents.
	More or less tangible and more or less emotional environmental, economic and social benefits: investments in order to reduce energy consumption during the use of the product could lead to an environmental impact which is lower than that which would be obtained by choosing other materials to optimise the features of the product in the final stage of its life cycle.
	Environment, technical aspects and quality: the use of some materials which can lead to benefits for the environment may undermine the reliability and the durability of the product.

It is important to identify additional possibilities for recycling the repulping residuals (the non‐fibre materials, i.e. polyethylene and aluminium, which remain after re‐pulping the package). An important multinational firm, operating in the packaging field, has, for example, developed a new recycling process based on plasma technology. This process separates these residuals to provide pure aluminium, which can be reused to produce new packages; and paraffin, which can be used in the plastics industry (Huat 2005). In recent years, there has been a growing demand for cartons that can be reclosed – the product lasts longer and less is wasted. However, more materials are required to make such packaging. Many firms have designed the innovative Flexi‐Cap that can be reclosed and, being made of plastic, is extremely lightweight and can be flat‐packed when empty, allowing for extremely efficient distribution to customers' filling plants.

The second matrix developed shows that the three most important ECs to consider in a DfE for filling machines are:

1. Replace some solvents. Emissions of volatile organic compounds (VOC) mainly arise from solvents used in printing inks and to some extent from printing plate production. There has been a downward trend in VOC emissions from firms over the last three years, which is due partially to a decrease in the production of packaging material requiring solvent‐based printing inks and to the installation of cleaning equipment.

2. Minimise the need for heating and cooling the machines. Several factories have established energy teams to ensure a focused and structured approach to energy saving. Sub‐monitoring systems are increasingly applied in order to facilitate energy management.

3. Reduce the product waste (fruit juice in this case study) during the filling process.

The design specifications chosen must be reviewed and analysed according to an environmental checklist which covers the entire life‐cycle of the product and considers: the choice of materials, technical optimisation of the production line, optimisation of the distribution system, optimisation of the product so as to attain better management in the final stage of the lifecycle.

It is necessary to highlight some aspects and problems that came out during this study:

	The panel of experts created for this work worked for a period of about two weeks, and the sessions were planned on a three‐round Delphi process;
	One of the most important problems in the procedure proposed is to create a large enough panel of experts (10 people) willing to invest time in this study;
	Moreover, initially it was not easy to find a common communication language since the group members came from various production sectors and also had different functional specialisations and skills.

6. Conclusions

Owing to the speed at which technology and knowledge in general evolve and because of the complexity and the variety of the products, it is not easy to define a general strategy for integrating environmental features into the design processes. This study investigated a new model for integrating environmental aspects into machine and machine component design. The first part of this paper showed how DfE principles can be introduced into all stages of machine development. In particular as far as the first stage, ‘Project Definition’, is concerned a virtuous circle has been set up based on two types of input:

	LCA of the entire supply chain;
	Analysis of legal requirements and guidelines for a DfE,

which by using QFD matrices lead to the definition of the main ECs, to be considered by designers in the subsequent stages of product development. These matrices also allow the machinery involved in manufacturing the product as well as the machine components to be defined.

The most important design specifications must be integrated according to the internal design standards of the firms. Current laws and regulations do not specify levels of environmental performance which must be reached. Therefore, on one hand it is more difficult for the firms to make comparisons with their main competitors, while on the other hand the firms involved have more freedom to act whatever the initial level of their ‘environmental maturity’ might be.

It is important to underline that many firms design and assemble components and modules for filling machines, and therefore most of the machine production and manufacturing process takes place outside the firm itself. Being aware of the importance of the environmental impact of the whole life‐cycle of the product means that it is necessary to try to control more carefully the stages at the start of the chain of values and then insert more levels of detail (level 2, level 3, etc.). It becomes important to ask for the collaboration of the suppliers and to carry out more and more frequent inspections to check up on their activities and their internal procedures.

This study, which allowed the definition of ECs connected with the most relevant environmental aspects of filling machines, is of general value and can be applied to all production systems involved in the beverage carton production line; from the machines used for the production of the raw materials to the machines which manufacture the roll of coupled material made up of wood‐fibre, polymers and aluminium.

Note

1. In the case studied the limit of the first class is 100/9–100/(9×2)≈5.5; while the limit of the second class is 100/9+100/(9×2)≈16.5. For example the EC of the first matrix (Figure 4) ’produce rectangular beverage cartons‚ has a relative importance of 7.6%. This value is included between 5.5% and 16.5% for this reason this EC has an importance of what is the second matrix (Figure 5) equal to 3.