Metrics for Measuring Sustainable Product Design Concepts

: Although products can contribute to ecosystems positively, they can cause negative environmental impacts throughout their life cycles, from obtaining raw material, production, and use, to end of life. It is reported that most negative environmental impacts are decided at early design phases, which suggests that the determination of product sustainability should be considered as early as possible, such as during the conceptual design stage, when it is still possible to modify the design concept. However, most of the existing concept evaluation methods or tools are focused on assessing the feasibility or creativity of the concepts generated, lacking the measurements of sustainability of concepts. The paper explores key factors related to sustainable design with regard to environmental impacts, and describes a set of objective measures of sustainable product design concept evaluation, namely, material , production , use , and end of life . The rationales of the four metrics are discussed, with corresponding measurements. A case study is conducted to demonstrate the use and effectiveness of the metrics for evaluating product design concepts. The paper is the ﬁrst study to explore the measurement of product design sustainability focusing on the conceptual design stage. It can be used as a guideline to measure the level of sustainability of product design concepts to support designers in developing sustainable products. Most signiﬁcantly, it urges the considerations of sustainability design aspects at early design phases, and also provides a new research direction in concept evaluation regarding sustainability.


Introduction
The world population increased from 1 billion in 1800 to 7.8 billion in 2020 [1]. Although the world population growth rate is decreasing, it is predicted that the world population will reach 9.7 billion by 2064, subsequently declining to 8.8 billion in 2100 [2]. The population growth has led to global issues, such as overconsumption of resources and energy and pollution. Sustainability is a fundamental attribute playing an increasingly significant role in design nowadays, especially in product design, where it is considered a requirement for solving those global challenges [3].
In recent years, there has been an increasing focus on sustainable product developments due to environmental regulations and expectations of consumers [4]. It is imperative for modern firms and enterprises to consider sustainability in product design and development [5]. Sustainable development is often defined as 'development that meets the needs of the present without compromising the ability of future generations to meet their own needs' [6]. Sustainability commonly involves three interconnected pillars: social, economic, and environmental [7]. However, sustainability in design is primarily focused on environmental aspects [4]. Therefore, sustainable design is described as a design approach for reducing the environmental impacts throughout a product's entire life cycle [8][9][10].  [16,20]). However, product design engineers generally focus on producing products that meet the required technical performance, aesthetics, durability, and cost demands [25], but lack awareness of the wider environmental impact of the design [12] (Knowledge Gap 1). For instance, Brundage et al. [26] indicated that there is a lack of communication between designers and manufacturers, which limits designers in reducing environmental impacts Figure 1. Opportunity in conceptual design for minimising cumulative 'lock-in' of environmental impact (adapted from [16,20]). However, product design engineers generally focus on producing products that meet the required technical performance, aesthetics, durability, and cost demands [25], but lack awareness of the wider environmental impact of the design [12] (Knowledge Gap 1). For instance, Brundage et al. [26] indicated that there is a lack of communication between designers and manufacturers, which limits designers in reducing environmental impacts from a manufacturing perspective. It is also challenging to improve a product's sustainability once the product is designed [21,22] (Knowledge Gap 2). Therefore, there is a need to project later sustainability-related activities, such as manufacturing, use, and end of life, to early design stages to inform better decision making [27][28][29]. For example, a product designed for easy manufacturing and assembly could increase the chances of it being reused or recycled, leading to a reduction in environmental impacts [30]. Nevertheless, the majority of existing concept assessment methods or tools are used to evaluate the feasibility and creativity of the concepts generated at early design stages [31][32][33][34][35][36][37] (Knowledge Gap 3). Therefore, current concept assessment methods could not guarantee the generation of sustainable product design concepts, which embody minimum environmental impacts. These three knowledge gaps imply that there is a need to come up with an approach to measure the sustainability of design concepts, considering aspects of later activities in the product life cycle and therefore promoting a more sustainable design manner.
The paper aims to offer support to designers in generating sustainable design concepts by considering a wide range of aspects to minimise negative environmental impacts, ultimately leading to sustainable products. The primary objective of the paper is to answer the following research questions: (1) What are the critical factors related to sustainable design, particularly environmental impacts? (2). Is there a set of metrics that can measure sustainable product design concepts?
The remainder of the paper is organised as follows: the next section reviews the related work on sustainability and concept assessment. In Section 3, four metrics (material, production, use, and end of life) for measuring sustainable product design concepts are proposed with rationales and measurements. A case study demonstrating the application of the metrics is provided in Section 4, followed by discussion in Section 5 and conclusion in Section 6.

Sustainable Product Design
Many methods and tools have been developed to support sustainable product design. Life cycle assessment (LCA) is most often used [38][39][40][41], which is a framework for calculating the environmental impacts of a product or service along its life cycle [42]. However, LCA methods are more suitable to be implemented at later design phases, as it often requires a large amount of data while the design of a product is more defined, of which materials, components, and processes are specified [43,44].
Quality function deployment (QFD) is another popular approach for supporting sustainable product design. For instance, Bereketli and Erol Genevois [45] and Younesi and Roghanian [46] employed quality function deployment for environment (QFDE) to consider both environmental and economic aspects. Wu and Ho [47] and Ocampo et al. [48] came up with integrated QFD approaches to address uncertainties, ambiguities, and interdependencies of decision parameters. Although QFD-based approaches provide benefits, such as considering the voice of customers and logical organisation of information, it could be complex and time-consuming to process a large matrix and challenging to offer valid quantitative information [49,50].
In addition to LCA-and QFD-based tools, several other types of methods have been developed to support sustainable product design, such as CAD-integrated tools, diagram tools, checklists, and guidelines, as well as Design for X approaches [42]. However, most of these tools require training and experience to implement, and some require specific knowledge to interpret results in order to gain insights. More importantly, very few existing sustainability tools or methods are found to be capable of supporting designers in sustainable product design at the conceptual design stage, concept assessment in particular, from the review conducted in this study.

Concept Assessment Methods
Concept assessment is often used to assess feasibility and creativity, and sometimes sustainability, aspects of design concepts in new product development. In this study, feasibility refers to whether a product produced could meet the design requirements, such as functional performance, business constraints, and customer needs. Creativity indicates the novelty, usefulness, and surprise of a design [51]. Sustainability refers to designing a product with minimum negative environmental impacts. Feasibility, creativity, and sustainability are not mutually exclusive; they can be interrelated when evaluating product design concepts. For example, the usefulness of a product is generally equivalent to the product's functional performance [51].
General multicriteria decision-making techniques are often used to assess design concepts, such as the Harris profile [31], Pugh matrix [34], and tabular evaluation matrix [37]. These methods often employ matrices consisting of a series of weighted criteria against which the design concept needs to be assessed. A ranking of the concepts with a quasi-quantitative measure of the advantages and disadvantages is provided to support the selection of the most suitable concept [37]. Although these methods could be used to assess creativity or sustainability depending on the criteria selected, they are mainly used to assess the feasibility of the concepts generated.
There also exist several other methods aimed at assessing the feasibility of design concepts generated. For instance, Davoodi et al. [35] applied the technique for order of preference by similarity to ideal solution (TOPSIS) to select the best concept based on the distance from ideal and non-ideal solutions. A shorter distance indicates a better design concept alternative, and vice versa. Goswami and Tiwari [52] suggested a framework to select the best design concept with the least risk by employing a Bayesian network methodology. Zhu et al. [15] and Tiwari et al. [13] proposed integrative methods by employing VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) to perform multicriteria decision analysis for selecting the best design concept. Shidpour et al. [14] indicated a multicriteria design concept evaluation method based on rough set theory for assessing the quantitative criteria of a product (such as costs) and fuzzy set theory for assessing the qualitative criteria (such as aesthetics). Rondini et al. [53] proposed a two-step importance-performance analysis (IPA)-based method, but focused on a product-service system. It is used for demonstrating the trade-off between customer value and provider value to direct design teams in progressing solution principles.
Creativity assessment methods aim at selecting the most creative concept and identifying creative designers and inventors [36]. The consensual assessment technique (CAT), proposed by Amabile [32], is known as the 'gold standard' in creativity assessment, which is grounded in the definition of creativity that 'the process by which something so judged (to be creative) is produced'. The CAT method often employs a group of experts in the domain in question to evaluate the creativity of a product.
However, human-judgement-criteria-based methods are most often used in product concept creativity assessment. For example, the Creative Product Semantic Scale (CPSS) [33,54,55] is a popular method for assessing design creativity. It involves three dimensions: novelty, resolution, and elaboration and synthesis, with associated subdimensions. Horn and Salvendy [56] indicated the use of novelty, affect, and importance, which are connected with consumer satisfaction, for assessing product design creativity. Sarkar and Chakrabarti [36] came up with a method for assessing creativity by evaluating the novelty and usefulness of a design concept. Novelty is assessed by using the SAPPhIRE model [57] and the function-behaviour-structure (FBS) model, while usefulness is assessed based on level of importance, rate of popularity of use, frequency of usage, and duration of use. Chiu and Shu [58] used novelty, usefulness, and cohesiveness to measure design concepts' creativity. Demirkan and Afacan [59] proposed three assessment factors related to the shape, characteristics, and design principles of a product, respectively. Lee et al. [60] measured design creativity by using novelty, usefulness, aesthetics, and complexity. Srinivasan et al. [61] employed novelty and quality, while Starkey et al. [62] used usefulness and uniqueness for assessing the creativity of product design concepts.
In comparison with the methods for assessing the feasibility and creativity of product design concepts, a limited number of methods have been developed to evaluate the sustainability of product design concepts. For instance, Lindow et al. [63] presented an interdisciplinary method based on the combination of the house of quality and life cycle sustainability assessment, considering both product properties and sustainability indicators. However, this method requires a large amount of information to perform a complete assessment, which is time-consuming and expensive. Hassan et al. [64] came up with a systematic approach to assess the sustainability of alternative part configurations. It employs a weighted decision matrix to determine the sustainability scores of design configurations and an artificial neural network to measure the sustainability performance. However, this approach can only assess the sustainability of a single part of a product rather than the whole product. Turan et al. [65] developed a sustainability assessment model in product development to support designers in making better decisions before completing the final concept. The model integrates a green project management concept for guiding the sustainability assessment, a new scale of weighting criteria for easing the rating process, and rough-grey analysis for supporting decision making. The assessment model is designed specifically for the automotive industry, limiting its application.
A summary of the concept assessment methods for product design illustrated in this section is presented in Table 1, with highlights of whether the method is aimed at assessing feasibility, creativity, or sustainability. As shown in the table, the majority of the concept assessment methods are aimed at evaluating the feasibility or creativity aspects of the concept generated. Sustainability aspects in concept assessment have been overlooked by most researchers, and therefore, it is potentially worthwhile to explore a set of metrics for measuring sustainable product design concepts. Table 1. Summary of concept assessment methods for product design.

Metrics for Measuring Sustainable Product Design Concepts
Four metrics, material, production, use, and end of life, for measuring sustainable product design concepts are proposed based on existing studies on sustainable design and conceptual design. The four metrics are described in the following subsections with the underpinning rationales and measurements, respectively. However, it is challenging to determine the actual value of the negative environmental impacts caused at the conceptual design stage. As a result, to reflect the level of sustainability in a simple but effective manner, measurement scales of low (0), medium (1), and high (2) are employed to indicate sustainability attributes.

Material
The assessment of materials in sustainable product design concepts is of fundamental importance. Materials have direct impacts on products with regard to the origin, property, and use of materials. Origin of materials refers to where the materials, used in the components and parts of a product, are originally sourced, involving nonrenewable and renewable resources. Nonrenewable resources are limited in supply, which cannot be replenished or replaced, such as fossil fuels, minerals, and metal ores. Popular materials produced from nonrenewable resources involve fossil-based plastics, metals, and glasses, which are often used in product design. Renewable resources refer to those that can be easily regenerated. Materials that are renewable, such as bamboo, mushroom, natural rubber, wood, and cotton, are increasingly used in product design. An increasing number of sustainable materials are being developed by applying renewable resources. For example, bioplastics, which are less or minimally reliant on fossil fuel, are produced by using renewable plants, such as sugarcanes, corns, and potatoes. In addition, there is currently an emerging trend towards utilising waste materials for design. For instance, by-product waste materials (such as chicken feathers and bran), which are secondary products generated from production, are often used as raw materials.
Toxicity, recyclability, and biodegradability are the main indicators of material sustainability properties. A material that is recyclable or biodegradable and nontoxic is often preferred, while materials that are toxic and neither recyclable nor biodegradable should be avoided in product conceptual design.
The use of material in a product refers to the volume/weight of materials and the number of types of materials involved. Using less volume/weight of materials contributes to a positive environmental impact, as it consumes less amount of resources and energies from material sourcing and production, to product end of life. The more types of materials used will increase a product's complexity, which will lead to more negative environmental impacts throughout the product's life cycle, as it increases the difficulties in product production and end of life. Therefore, determining which material(s) to use and identifying how the material(s) are used in a product design concept are strongly associated with the product's sustainability performance.

Measurement of Material
In the conceptual design stage, information such as material origins, material properties, and the use of materials needs to be determined to assist designers with evaluation. As suggested previously, low (0), medium (1), and high (2) are used to indicate sustainability attributes. For example, if the origin of one type of material used is a promising renewable source, then a rating of high, a score of 2, will be given. Similar principles apply to material properties. For example, if the material is toxic, cannot be recycled, or is biodegradable, then a rating of low, a score of 0, will be given. The use of materials includes the weight/volume of materials and the number of types of materials used. Regardless of material origins, lesser volume/weight and fewer types of materials used will lead to less negative environmental impacts. However, it is difficult to judge the absolute quantity (weight/volume) of materials needed for a concept; therefore, this attribute refers to the potential for material quantity reduction at the time when the concept is evaluated.
For instance, if the volume/weight of materials of a product concept could be easily reduced without affecting the structure and performance of the product, a high (2) score will be given. Table 2 summarises the attributes to consider for material sustainability and provides a brief explanation of each level to inform rational decision making. An equation is then developed to quantify the material sustainability, as shown in Equation (1). In Equation (1), i refers to the ith type of material used in a concept. A multiplier is used to correlate attributes that have aggregated effect. For example, material original (M 1 ) and use of material-quantity (M 3 ) have a clear aggregated effect; hence, they are multiplied together. These aggregated effects are added together and then divided by the number of material types N to indicate the overall material sustainability, which will vary between 0 and 8. In order to yield an accessible result, a scaling process is performed to ensure that the final score is within the range of 1 to 10, in which 1 means poor and 10 means excellent with regard to sustainability.

Production (Manufacturing and Assembly)
Producing products in a sustainable manner, such as conserving resources, consuming less energy, and generating less pollution and waste, leads to minimum negative environmental impacts. However, production is a complex process where many design details are determined at the detailed design stage rather than the conceptual design stage. Therefore, only aspects related to manufacturing and assembly are discussed in this paper. Design for manufacturing and assembly (DFMA) is an effective approach to achieve sustainable production. This study extracted core DFMA considerations, for ease of assembly and manufacturing, to measure sustainable production aspects of design concepts, as shown in Table 3. Minimising the number of parts in a practical manner, as well as using more standardised parts/components and fewer unique parts/components, could reduce inventory cost, process time, and so on. Designing parts for ease of assembly involves better presentation (such as avoiding too large or too small items and employing symmetric features), easy handling (such as avoiding oversize, sharp, slippery, heavy, and fragile items), mistake proofing (such as using symmetric or asymmetric features to prevent parts from being assembled in wrong orientations), and efficient insertion (such as employing self-aligning/locating features). Suitable fabrication methods refer to the identification of the most appropriate technology/process based on the material selected to minimise excessive operations, such as polishing and fine machining.

DFMA Considerations Explanations
Minimum number of parts The practical minimum number of parts for both manufacturing and assembly.
Parts/components standardisation The number of standardised parts/components and unique parts/components.
Parts assembly Parts presentation, handling, mistake proofing, and insertion for ease of assembly.
Suitable fabrication methods Cost-and energy-effective technology/process for ease of manufacturing.

Measurement of Production
Production details, such as manufacturing methods, manufacturing parameters, and assembly procedures, can be difficult to determine at the conceptual design stage. However, designers are encouraged to consider these attributes with respect to sustainability to steer towards a more sustainable outcome. A similar approach explained in 3.1.1 is adopted here, for example, a low rating, a score of 0, will be given if a concept requires a considerable number of customised parts/components, implying that more negative impacts are created during production. An explanation of levels for each production attribute is provided in Table 4. It is worth noticing that these production attributes are concept dependent, and this therefore requires subjective judgement, for instance, whether adequate standardisation has been achieved. In addition, unlike the types of materials, it is also challenging to evaluate each individual part/component with respect to standardisation, fabrication, and assembly; therefore, here they are considered holistically at the concept level. An equation is then developed to quantify the sustainability with respect to production, as shown in Equation (2). Similar to material, the aggregated effect for production attributes is considered here. This is again reflected on the multiplier. For example, balance between number of parts and their complexity (P 1 ) is closely related to part standardisation (P 2 ), while suitable fabrication method (P 4 ) will amplify their effect, hence, P 1 × P 2 × P 4 . The equation leads to an overall score for production sustainability, which will vary between 0 and 12. A similar scaling process is performed to ensure that the final rating will yield a value between 1 (poor) and 10 (excellent).

Use
The use of a product pertains mainly to the amount of time the product is owned and operated by its user. A product's lifetime starts from when it is acquired to when the product is discarded, which is primarily determined at the conceptual design stage. Functional obsolescence, maintenance prevention, and aesthetic obsolescence are the main reasons that lead to the end of life of a product. In early studies, product lifetime extension was employed to reduce resource consumptions and waste productions by means such as ease of repair and upgrade. However, a longer life span of a product does not necessarily indicate that the product is more resource and waste efficient. For instance, longer lifetime products usually consume more resources in material and production. These extra resources are wasted if a product's lifetime is longer than the time of the product being needed by the user. Therefore, product lifetime optimisation, where a balance between extending and shortening the lifetime and use time is achieved, should be used as an effective strategy to minimise the negative environmental impacts of products. For example, less durable materials should be used for short-life or temporary products and parts. Another strategy to decrease environmental impacts at the use stage is to reduce the product's resource or energy consumption. For example, LED lights consume much less electrical energy in comparison with incandescent lights, but produce the same illumination. Therefore, LED lights should be used rather than incandescent lights while designing products with illumination features.

Measurement of Use
As described previously, the balance between product use time and lifetime needs to be considered during the conceptual design stage. An ideal scenario would be when the product use time is identical to its lifetime, implying that the product enters its end-of-life stage immediately after the use stage. Therefore, the ratio between product lifetime and use time is an attribute to consider, as shown in Table 5. Despite various product categories being evaluated, the product use time and lifetime balance can be determined in a unified way, meaning that the difference between them should always be minimised. For example, the perfect balance for a disposable coffee cup is that it can be recycled right after people finish their drinks. For a mobile phone, the ideal case would be that it can be recycled right after it breaks or when people get a new one rather than sitting in a drawer. It is possible to use objective values to determine the thresholds (low, medium, and High) of product use time/lifetime (U 1 ), but they would be largely dependent on the products themselves. Therefore, subjective descriptors, such as 'significantly shorter/longer', are employed. The concept requires a significant amount of resource to maintain/service The concept consumes a fair amount of resource to maintain/service The concept consumes a slight amount of resource to maintain/service Energy consumption during use directly indicates the energy efficiency; hence, it is used as the second attribute. Different products can vary significantly; hence, it would be difficult to judge without considering the product category. As a result, it would be beneficial to develop a lookup table by collecting data of day-to-day products and come up with a range of specific values for energy consumption for different product categories. By this, the designer could make judgements by referring to the table more easily. However, it is time-consuming to construct such a lookup table, and therefore, subjective descriptors are used for energy consumption during use (U 2 ) in this study. Robustness, reliability, and maintenance are treated as the third attribute to indicate sustainability. For example, if a design is more robust, reliable, and easy to maintain, it is then unlikely to cause a significant negative environmental impact due to malfunctioning and servicing. Equation (3) is then developed to quantify the use sustainability. Again, the aggregated effect of attributes is considered here and denoted by multipliers. For example, the effect of energy consumption during use (U 2 ) will be amplified by the product use time/lifetime (U 1 ), hence U 1 × U 2 in the equation. The same scaling process is applied to ensure that the final sustainability score falls between 1 and 10.

End of Life
End of life refers to a product that is at the end of its life cycle, where the product needs to be discarded. End-of-life approaches, such as recycling, reuse, repair, and remanufacturing, are considered more sustainable than conventional disposal methods involving incineration and landfill. Employing biodegradable materials and using waste-to-energy technologies are often used to decrease the negative impacts caused by product disposals, such as landfill and incineration. However, product disposal still leads to issues, such as pollutions and contaminations, and therefore is considered unsustainable. Recycling is a process of converting a disposed product into new materials or objects; reuse involves the action of using the product or parts of the product, without changing the structures, for original and new purposes; repair refers to the replacement of nonfunctional or damaged parts of the product; and remanufacturing means returning the product to a 'like-new' condition. Product disassembly is often needed and considered a significant process in product end of life, even for landfill and incineration. Ease-of-disassembly tactics, such as employing detachable joints, using standardised fasteners, minimising the number of fasteners, and avoiding glues, should be considered at the conceptual design stage to contribute to sustainable product end of life. In addition to ease of disassembly, strategies such as using compatible materials, employing modular parts, ease of identification and inspection, and ease of sorting could also support product end-of-life processing for better environmental performance.

Measurement of End of Life
Compared with recycling, remanufacturing, and repair, reuse requires the least resource and, therefore, is listed as an individual attribute. Recycling, remanufacturing, and repair all require further handling and processing, which consumes more energy and materials; hence, they are categorised together. Some parts of a product are inevitably not reusable, recyclable, remanufacturable, or repairable and need to be disposed of. As a result, the environmental impact caused by disposal needs to be considered. A product at its end of life often requires disassembling to obtain the parts to be reused, recycled, remanufactured, repaired, or even disposed of. Therefore, the degree to whether the concept is easy to disassemble at its end of life is another important attribute. Table 6 presents a summary of the explanations for the attributes discussed. Similar to other metrics, the potential aggregated effect is represented by multipliers of attributes; for example, in order to reuse (E 1 ), recycle, remanufacture, and repair (E 2 ) and dispose (E 3 ) the components of a product, the ease of disassembly (E 4 ) of the product is critical. Equation (4) with a scaling operation was developed for the end-of-life sustainability.

The Four Metrics
Material, production, use, and end of life are the four metrics, involving 15 attributes, proposed for measuring sustainable product design concepts. A summary of the metrics is depicted in Figure 2. The four metrics proposed could be used individually to measure specific aspects of a product design concept's sustainability, and integrated to provide insights into the concept's overall sustainability. Equations (1)-(4) are developed to indicate the degree of sustainability, from poor (1) to excellent (10). A demonstration of utilising the four metrics in a systematic manner for measuring sustainable product design concepts is presented in Section 4. is depicted in Figure 2. The four metrics proposed could be used individually to measure specific aspects of a product design concept's sustainability, and integrated to provide insights into the concept's overall sustainability. Equations (1)-(4) are developed to indicate the degree of sustainability, from poor (1) to excellent (10). A demonstration of utilising the four metrics in a systematic manner for measuring sustainable product design concepts is presented in Section 4.

Case Study
A case study was conducted to demonstrate the application of the metrics explored. Two design concepts of portable blenders, as shown in Figure 3, were used in the case study for sustainability measurement by a design expert by employing the four metrics, material, production, use, and end of life, and associated measurements illustrated previ-

Case Study
A case study was conducted to demonstrate the application of the metrics explored. Two design concepts of portable blenders, as shown in Figure 3, were used in the case study for sustainability measurement by a design expert by employing the four metrics, material, production, use, and end of life, and associated measurements illustrated previously. The two blender concepts were generated, respectively, by two novice design engineers, who participated in the case study voluntarily with high levels of interest. To be more specific, the two design novices were asked to come up with a conceptual design of a portable blender and present the concept using sketches or CAD drawings with annotations. The key product design specifications of the portable blender design were provided to the two design novices to guide them in the conceptual design stage, as shown in Table 7. The design novices signed up with standard case study protocols to provide permission to use their design concepts in this publication and related analysis.

EER REVIEW
13 of 20 a portable blender and present the concept using sketches or CAD drawings with annotations. The key product design specifications of the portable blender design were provided to the two design novices to guide them in the conceptual design stage, as shown in Table 7. The design novices signed up with standard case study protocols to provide permission to use their design concepts in this publication and related analysis.

Function
• To process nuts, fruits, and vegetables into liquid. • The blender should be able to operate without plugging into a power socket. Performance • The needed operating time should be within 30 seconds.
• The residual particles should not exceed 3 mm in diameter.
• The blender needs to be easy to clean. Safety • The blender should not possess risks of causing injuries at any time.

Purpose market
• For outdoor use (e.g., travel, hiking, and picnic). Quantity • For product trial, 300 units are expected. Quality and reliability • Food hygiene standard has to be met.
• The blender should withstand high/low pressure, shock, dust, and water. Cost • Should not cost more than a conventional juice blender. As shown in Figure 3, concept (a) is a portable blender powered by a 3 V DC motor using two AA batteries. The body part of the blender consists of a jug section and a base section, but they are not detachable. The base section is used to house the blade, motor, AA batteries, and other associated electronic components, while the jug section is for containing ingredients, such as fruits and vegetables. A cap is designed with a spout lid, which is screwed on the jug during blending and drinking. PET (polyethylene terephthalate) was selected to produce the cap, jug, and base, while injection moulding was selected as the manufacturing process. Stainless steel was selected for manufacturing the blade via casting. Concept (b) is a manually powered handheld design without the need for batteries or AC/DC power supply. It consists of a lid and a body section. There are two lids: One is for sealing the container body properly while providing interface for attaching the handle bar. Another lid can be used when the blending is finished, and it has an aperture for easy dispensing. The body section contains a transparent PET container with stainless steel shaft and two sets of blades. The steel shaft and blades can be removed completely for easy cleaning and replacement. Table 7. Key product design specifications. Function • To process nuts, fruits, and vegetables into liquid.

•
The blender should be able to operate without plugging into a power socket.

Performance
• The needed operating time should be within 30 s.

•
The residual particles should not exceed 3 mm in diameter.

•
The blender needs to be easy to clean.

Safety
• The blender should not possess risks of causing injuries at any time.

Quantity
• For product trial, 300 units are expected.

Quality and reliability
• Food hygiene standard has to be met.

•
The blender should withstand high/low pressure, shock, dust, and water.

Cost
• Should not cost more than a conventional juice blender.
Size and weight • The blender should be easy to carry with one hand.

•
The weight should not exceed 2 kg.
Life span • The blender should last at least 6 months with daily use.

Recycle
• The blender should be easy to recycle at its end of life.

Environment
• The blender should cause minimum environmental impact across its entire life cycle (e.g., service life and end of life).

Evaluation and Results
The design expert, with over 8 years of experience in design engineering, participated voluntarily in the case study to assess the sustainability of the two portable blender concepts. Prior to starting the assessment, explanations of the metrics and instructions for using the equations were provided to the design expert. The results of the evaluation are shown in Table 8. For instance, for the material metric of concept (a), four types of materials, PET, stainless steel, battery, and motor, were included in the evaluation. PET was the material used for the cap, base, and jug, while stainless steel was used for the blade of the blender. The motor and battery are not strictly materials, as they have multiple components and features involving many types of materials. However, it is challenging, as well as time-consuming, to consider all materials used for fabricating a motor and battery. Therefore, components such as a motor and battery, which contain multiple materials, are considered individual types of materials to ease the evaluation process. For the material origin (M 1 ) attribute, PET, stainless steel, battery, and motor were all produced using nonrenewable resources, such as fossil fuels and minerals, and therefore, low scores of 0 were given. For the material property (M 2 ) attribute, PET is a recyclable material but is toxic to the environment; thus, a medium (1) score was given. Stainless steel is a nontoxic and recyclable material, therefore achieving a high (2) score. Both the battery and motor contain materials that are nonrecyclable and toxic, therefore receiving low (0) scores. For the use of material-quantity (M 3 ) attribute, both PET and stainless steel have fair potentials for material quantity reduction, such as reducing the thickness of the wall or the size of the blade, and were therefore assigned medium (1) scores. There are poor potentials to reduce material quantities for both the battery and motor; thus, low (0) scores were given. As a result, concept (a) achieved a material metric score of 1.84 according to Equation (1). The measurements of the other metrics of concept (a), as well as the sustainability measurement of concept (b), are depicted in Table 8. Table 8. Evaluation of the two blender concepts.

Material
Material origin (M 1 ) Use of material-quantity (M 3 ) Use of material-type (N) 4 4 Material metric Production Balance between the number of parts and complexity (P 1 ) Simple design, only a few simple parts-2 Simple design, only a few simple parts-2 Parts standardisation (P 2 ) Motor, battery, and gearbox can benefit from standard components-2 All components require customisation-0 Parts design for assembly (P 3 ) Good potential for assembly optimisation-2 Good potential for assembly optimisation-2 Internal components for the base (e.g., motor and gearbox will require a fair amount of resource to maintain/service)-1 All components are easy to maintain-2 Use metric

End of Life
Reuse (E 1 ) Battery, motor, and gearbox have fair potential to be reused-1 Poor potential for reuse as all components are custom-made-0 Recycling, remanufacturing, and repair (E 2 ) PET plastic, steel blades, and batteries can be recycled (i.e., fair potential for recycling and remanufacturing)-1 PET plastic, steel shaft, and blades can be recycled; bamboo can be remanufactured-2 Disposal (E 3 ) Batteries and motors will cause moderate impact due to disposal-1 All components will have little impact due to disposal-2

Ease of disassembly (E 4 )
Blender base that contains battery, motor, and gearbox will be difficult to disassemble-1 All components are easy to remove-2 An overview of the product concept sustainability assessment results of concepts (a) and (b) is shown in Figure 4, including the results of the four metrics and the average sustainability score. As shown in the figure, concept (a) achieved 1.84 in material metric, 10 in production metric, 5.5 in use metric, and 3.25 in end-of-Life metric, with an average sustainability score of 5.15. Concept (b) achieved material, production, use, and end-of-life metric scores of 2.69, 2.5, 10, and 7, respectively, with an average sustainability score of 5.55. Based on the results, although the average sustainability scores of concepts (a) and (b) are at a similar level, their metric scores are different. For example, concept (a) outperforms concept (b) in terms of production, as it employs standard components and requires no excessive fabrication operations, as presented in Table 8. However, concept (b) has higher scores in material, use, and end-of-life metrics in comparison with concept (a) due to features such as using renewable and less toxic materials, employing manual operation, and being easy to maintain, recycle, remanufacture, and disassemble. features such as using renewable and less toxic materials, employing manual operation, and being easy to maintain, recycle, remanufacture, and disassemble. In addition to the comparison, the sustainability assessment results also indicate sustainability improvement directions for each concept. For example, concept (a) achieved poor performance in the material metric, especially the material origin (M1) attribute, as shown in Table 8. Therefore, using renewable materials, such as recycled glass, to replace the use of PET for producing the jug of the blender could potentially increase the sustainability score of the concept regarding the material metric. Concept (b) has poor performance in production, of which the parts standardisation (P2) attribute is given a low (0) score, and the suitable fabrication method (P4) attribute is given a medium (1) score. Therefore, using a standardised rotating handle with shaft that is available on the market will reduce the number of customisation parts and the excessive fabrication operations needed, which could potentially improve the production score. Other improvement strategies could be inferred to increase the scores of the remaining sustainability metrics of the two concepts.

Discussion
Four metrics, material, production, use, and end of life, are proposed in this study for measuring sustainable product design concepts. The corresponding attributes with associated measurement equations could be used to identify the sustainability level of a concept, with regard to the four metrics, in a quantitative manner. The attribute scores, low (0) to high (2), applied are in the simplest form possible to provide the most straightforward indication of attribute sustainability. The equations developed aim to indicate the sustainability of each metric by using multipliers to link attributes that have aggregated effects, for instance, material origin and use of material-quantity, and using summation to indicate the cumulative effects of the attributes. The equations developed and used are not necessarily the final forms, and modifications can be envisaged. For example, different weights can be assigned to attributes within a metric, referring to different product design In addition to the comparison, the sustainability assessment results also indicate sustainability improvement directions for each concept. For example, concept (a) achieved poor performance in the material metric, especially the material origin (M 1 ) attribute, as shown in Table 8. Therefore, using renewable materials, such as recycled glass, to replace the use of PET for producing the jug of the blender could potentially increase the sustainability score of the concept regarding the material metric. Concept (b) has poor performance in production, of which the parts standardisation (P 2 ) attribute is given a low (0) score, and the suitable fabrication method (P 4 ) attribute is given a medium (1) score. Therefore, using a standardised rotating handle with shaft that is available on the market will reduce the number of customisation parts and the excessive fabrication operations needed, which could potentially improve the production score. Other improvement strategies could be inferred to increase the scores of the remaining sustainability metrics of the two concepts.

Discussion
Four metrics, material, production, use, and end of life, are proposed in this study for measuring sustainable product design concepts. The corresponding attributes with associated measurement equations could be used to identify the sustainability level of a concept, with regard to the four metrics, in a quantitative manner. The attribute scores, low (0) to high (2), applied are in the simplest form possible to provide the most straightforward indication of attribute sustainability. The equations developed aim to indicate the sustainability of each metric by using multipliers to link attributes that have aggregated effects, for instance, material origin and use of material-quantity, and using summation to indicate the cumulative effects of the attributes. The equations developed and used are not necessarily the final forms, and modifications can be envisaged. For example, different weights can be assigned to attributes within a metric, referring to different product design applications, to better indicate concept-specific sustainability. The major advantage of using equations is making quantitative comparisons between different concepts. The designer is able to obtain instant results of concept sustainability based on the scores. More significantly, scores for different metrics allow an indication of sustainability improvement directions. For instance, if a product design concept received a poor sustainability evaluation rating for a metric, the designer could explore which attribute(s) of the metric has low ratings and then modify the design concept accordingly for sustainability improvements.
This study has three implications. First, the four metrics identified for measuring product design concepts could effectively improve a product's sustainability level at a lower cost in comparison with addressing sustainability issues at later stages. Solving sustainability issues once a product is designed or at late design stages is challenging and expensive [21,22], whereas decisions made at the conceptual design stage have high impacts on minimising the negative environmental impacts on downstream activities, such as material, production, use, and end of life [13][14][15][16]. The second implication is that the study has raised the significance of addressing sustainability issues at the conceptual design stage. Many design features related to product sustainability or environmental impacts are not often considered by design engineers, as they generally focus on cost, performance, and durability [12,25,26]. This could result in products' lack of sustainability considerations, while the introduction of the four metrics for assessing the sustainability of product concepts has the potential to foster designers in considering sustainability design features during conceptual design. The third implication is the need for more and better sustainability concept assessment tools. The review conducted in this study reveals that the majority of existing concept assessment methods or tools are aimed at assessing the feasibility or creativity aspects of product design concepts. Although a few methods exist for evaluating the sustainability of product design concepts, these methods are limited in use [63][64][65]. The metrics for measuring sustainable product design concepts proposed in this study have an extensive application scope in practice, which could also be utilised as a theoretical foundation for developing advanced sustainability concept evaluation tools.
However, the metrics proposed are aimed at recommending design changes at the conceptual design stage, and it might be challenging to suggest final determinations. It therefore requires the designer or evaluator who uses the metrics for evaluation to possess sufficient knowledge and experience in sustainable design and decision-making skills to yield final design decisions. Further explorations, such as conducting more practical case studies, are needed to examine how well the four metrics represent sustainability to increase the metrics' suitability for conceptual design and to improve the measurement equations.

Conclusions
Sustainability plays an increasingly significant role in modern product design and development, while it is indicated that most of the negative environmental impacts are determined at early design stages, such as conceptual design. However, the review of prior literature showed that most of the existing concept evaluation methods are geared to measure the feasibility or creativity of concepts generated rather than the sustainability. The lack of measurements of sustainability at the conceptual design stage often leads to nonsustainable products, which result in negative environmental impacts. Therefore, this paper explored the key sustainable design elements and propose a set of metrics for measuring sustainable product design concepts. The four metrics identified, material, production, use, and end of life, associated with corresponding attributes and measurement equations, can support designers in producing sustainable design concepts, ultimately leading to sustainable products with minimal negative environmental impacts. Although this paper aims at assessing the environmental aspects of sustainable product design concepts, products produced also impact both the social and economic dimensions of sustainability through manufacturing, use, and end of life, which contribute to both employability and value creation [66].
The paper is the first study to explore metrics for evaluating product design sustainability at the conceptual design stage. It delivers three significant contributions to engineering design, sustainability, and energy research communities. First, it serves as a guideline to measure the level of sustainability of design concepts for supporting sustainable product design in a quantitative manner. Second, it urges design practitioners and researchers to look into the importance of considering sustainable design aspects at early design stages. Finally, the study offers new research insights into exploring sustainable concept evaluation and can be used as an infrastructure to develop future concept evaluation tools.