Limits of ecodesign: the case for open source product development

Abstract

In pursuit of coping mechanisms for the challenges of sustainable manufacturing, systematic methods have been developed to allow engineers to improve the ratio between products’ utility and environmental impact. These necessary efforts remain however limited if the surrounding production and consumption practices are not called into question at the same time. This article introduces open source product development as a potential approach for unlocking the limits of existing sustainable product design practices and to lead towards alternative and eco-efficient production and consumption practices termed as participative production. First, it discusses the potential advantages of this emerging practice in terms of environmental sustainability. Second, it provides a screened qualitative environmental analysis of 18 exemplary open source hardware products. Specific sustainable design principles implemented by these products are highlighted and discussed in order to identify challenges for further common research in the field of open source and sustainable product development.

Keywords:

• Open source innovation
• open source hardware
• sustainable product design
• eco-design

Beyond eco-design

Eco-design, the approach aimed at improving the ratio between the usefulness and environmental impact of a good or service, constitutes a necessary means of addressing the challenges of sustainable manufacturing (Stark, Seliger, and Bonvoisin 2016). Extensive research has been conducted since the 90s in order to develop appropriate design engineering methods, which contributed to the emergence of a large catalogue of more than 100 different eco-design methods and tools (McAloone and Pigosso 2017). Though necessary, this approach focused primarily on the product may be limited if the surrounding production and consumption practices remain unaddressed at the same time. Improvements in terms of environmental impact per product unit are limited by several constraints applying to product development in terms of performance, production cost, aesthetics, and standardisation, among other factors (for a more exhaustive list of these constraints, see for example Lagerstedt and Luttropp 2006). A strictly product-centred approach may not allow achieving more than incremental improvement in the environmental footprint where however radical cuts in environmental impact are required. Maintaining the pace of progress in eco-design approaches is one necessary matter, with however limited overall effect. Solutions which tackle sustainability on a more all-encompassing level are required – that is, solutions which not only call into question the technical design of the product, but also the production and consumption patterns behind the product. This aspect is one of the key propositions of the concept of Product Service Systems (PSS), a promising complementary approach which has been under the focus of the last decade. PSS are integrated offers of products and services focused on customer satisfaction rather than on exchanging product ownership (e.g. Tukker and Tischner 2006). While it is meanwhile understood that PSS are not inherently eco-efficient, this concept has been identified as a ‘potential sustainable business model’ because it may help in decoupling profit and production volume without constraining usage volume (Bocken et al. 2014).

Following the same logic, the present contribution discusses the potential of open source product development (OSPD) for leading to eco-efficient production and consumption patterns. In a first section, the concept of OSPD is introduced. The context, the specific characteristics and the challenges for further development of this emerging practice are described. OSPD is presented as a trigger of participative production whose implications in terms of environmental sustainability are discussed. This theoretical discussion is followed by an empiric study aimed at identifying synergies between current practices of OSPD and sustainable product design. It reports the study of the specific sustainable design principles implemented by 18 open source hardware products aimed at participative production. Finally, these principles are discussed as challenges for further common research in the field of open source and sustainable product development.

OSPD – definition and theoretical implications

Context of emergence

The spread of information, communication technology, and cheap small-size production tools like 3D-printers has enabled the emergence of the ‘maker movement’ (Voigt, Suero Montero, and Menichinelli 2016) based on the widespread participation of the individual citizen in the development and production of tangible products. A new category of ‘home engineers’ emerges which is supported by innovative businesses like Thingiverse¹ or Shapeways² providing online CAD-model sharing and remote 3D-printing services, hence allowing makers up- and downloading 3D models and getting physical prints by the mail. These services are two examples of numerous online design repositories supporting exchange of best practices and do-it-yourself assembly manuals between makers. In parallel, the easiness of sharing digital content also makes it possible for product development projects to publish online information so that their product can be physically replicated or further developed by spontaneously emerging online communities, such as in the case of the well-known RepRap 3D-printer.³ Product development teams may also emerge online out of the common willingness of users to shape solutions adapted to their needs. This is the case in the project Apertus Axiom,⁴ dedicated to the development of digital cinema production tools. Companies may also use the dynamics of online communities and open source product definition for setting up innovative business models like in the case of the company Inventables^TM,⁵ which distributes open source desktop-size CNC carving machines. All these initiatives have in common to be based on open source hardware, i.e. ‘hardtware whose design is made publicly available so that anyone can study, modify, distribute, make, and sell the design or hardware based on that design’ (Open Source Hardware Association, 2016). Existing examples of complex open source hardware products have been reported to cover product categories such as agriculture machinery, machine-tools, means of transportation, renewable energy supply technologies or even medical equipment (Bonvoisin, Mies et al. 2016). The current emergence of open source hardware shows great potential for product innovation and incubation of new businesses, and further, allows OSPD to be seen as potential ‘billion dollar business’ just as open source software has achieved in the last 40 years (Fjeldsted et al. 2012).

Definition and characterisation

In the context of this article, we define OSPD as the process of developing open source hardware. Just like open source software development, OSPD is a form of open source innovation, defined by Raasch, Herstatt, and Balka (2009) as the ‘free revealing of information on a new design with the intention of collaborative development of a single design or a limited number of related designs for market or non-market exploitation’. Huizingh (2011) differentiates between product development projects depending on what is being made open: either the development process itself (i.e. by giving the opportunity of every interested person to participate) and/or its outcome (i.e. by publishing product information). Balka, Raasch, and Herstatt (2014) further refine the concept of openness as a gradual and composite concept determined by three factors: (1) transparency – access to sufficient information to understand the project details, (2) accessibility – the possibility for community members to take an active part in the development, and (3) replicability – the possibility for self-assembly of the product. Empirical observations of OSPD projects have shown certain heterogeneity in the perception of the concept of open source in practice and confirmed the role played by these three concepts (Bonvoisin et al. 2016). Indeed, while the publication of product-related information is at the heart of the definition of open source hardware, motivations behind publishing this information may diverge. One motivation is in obtaining feedback on the design and in encouraging the emergence of collective intelligence, as expressed by the famous quote from Eric Raymond ‘given enough eyeballs, all bugs are shallow’ (Raymond 1999). Another impetus is to broadcast an innovation by allowing others to replicate and distribute the product.

In addition to their motivations, OSPD projects may be characterised by their effective level of openness. To be sure, products are not either completely open or closed, but in practice at least partially opened up. Companies may, for example, choose to publish some parts about their products’ design in order to profit from the dynamics of the crowd, while keeping some parts protected in order to safeguard key competences (Balka, Raasch, and Herstatt 2010). Additionally, and in comparison to Open Source Software, the ‘source’ of physical products may be more diffuse and also distribute information. Sharing CAD files may, for example, not be sufficient to reproduce a product. Some auxiliary information such as a bill of materials, a description of part requirements and material properties as well as detailed assembly instructions may be required. Due to capacity limitations or as a result of strategic choices, these documents may be partially released and their quality may vary.

Another differentiation characteristic is the size of the active community. A kind of continuum exists between a one-man or one-company project whose documentation is made open source, and a large community built around the development of a product from scratch. The concept of community itself is complex as members may have different levels of implication in the process. Their roles may range from just using the product and giving feedback to participating in the product development or even management activities. Different models have been proposed for describing members’ motivations and the structure of communities. Heyer and Seliger (2012) assume three groups generally distributed in a 1-9-90% proportion and distinguished by their degree of activity: (1) the deeply involved core team of developers who takes an active part in generating content and organising the development process (2) a pool of contributors who may generate content in a rather sporadic way and (3) a larger audience who follows the development of the project and is interested in some way utilising the results without taking part actively. Janzik (2015) considers not only the frequency of activity but also the reciprocity of the activities performed and defines five types of contributors: innovators, activists, tourists, fellow travellers and free riders (Janzik 2015). Ehls and Herstatt add a parameter to this complex picture by underlining the fickle nature of community involvement, as members may change their activity rate and switch types over time (Ehls and Herstatt 2015).

Finally, these characteristics may shift or morph over time alongside the project timeline. Some projects may, for example, start as a local project, go open and community-driven and then turn back to partially closed, as the famous example of MakerBot^TM,6 illustrates.

Challenges for further expansion of this emerging practice

As a disruptive alternative to current practices, OSPD raises numerous challenges for practical and systematic implementation. The exhaustive compilation of these challenges and of their implications for research and practice is not within the scope of the present article; however some salient aspects may be worth mentioning. In contrast to conventional industrial product development, contributions of project members’ are no longer sealed by contractual agreement but instead are voluntary. Hierarchical organisation and task assignment are replaced by a low level of restrictions, self-motivation and self-selection of tasks (Müller-Seitz and Reger 2010). In this context, the sequentially structured step-by-step approach of product development is not applicable anymore, and in its place, a more iterative and trial-and-error approach may come about. To date, the emergence of OSPD still suffers from a limited availability of adapted process support to assist in contending with the organisational challenges associated with collaboration work on the part of a diverse and dispersed non-contractually involved volunteer team (Bonvoisin and Boujut 2015). A large part of OSPD projects remains restricted to the development of products of low complexity and quality, i.e. prototypes or toys for do-it-yourself hobbyists (Hansen and Howard 2013) which cannot compete with industrial quality standards. Beyond the mere challenges of product development, basing successful businesses on OSPD also requires dealing with quality insurance and the associated legal risks. Moreover, communities may have to deal with the fact that not every form of participation is motivated by good intentions: the non-exclusivity of the participation in the development process requires protection mechanisms against vandalism, as experienced by Wikipedia (Adler et al. 2011) or other participative projects (e.g. Rahwan 2014).

Potential in terms of environmental sustainability

First, open source has been presented as a necessary enabler for scaling of Appropriate Technologies, i.e. ‘technologies that are easily and economically used from readily available resources by local communities to meet their needs’ (Pearce et al. 2010). In other words, appropriate technologies focus on empowerment and self-determination as well as on compatibility with the local environment. Second, while the concept of OSPD concerns primarily the participation of the customer and citizen in a virtual and intellectual process (that is, the product development process), it goes hand in hand with a parallel evolution of the role of the customer and citizen in production. This redistribution of roles between manufacturing industries and individuals have been alternatively termed ‘commons-based peer production’ (Benkler and Nissenbaum 2006), ‘personal fabrication’ (Gershenfeld 2007), ‘direct digital manufacturing’ (Chen et al. 2015), ‘bottom-up economy’ (Moritz et al. 2015), ‘redistributed manufacturing’ (Moreno and Charnley 2016), ‘distributed economy’ (Johansson, Kisch, and Mirata 2005), ‘do-it-yourself production’ (Bonvoisin, Prendeville, and Krishna Galla 2017), under the motto ‘design global, manufacture local’ (Kostakis et al. 2015). Kohtala (2014) summarised these concepts under the term ‘distributed production’ and demonstrated that their potential in terms of environmental sustainability raised significant interest in the scientific community. She showed at the same time that, unfortunately, little empirical research has been performed so far: publications focussed on distributed production and environmental sustainability are for a large part propositional.

In the following, the article concentrates on participative production as a specific subset of distributed production characterised by the participation of the individual citizen in the design and production of products. The following hypotheses regarding the environmental sustainability of participative production can be formulated:

•	H1 – Participative production supports longer service lives due to the promotion of robust product design. Indeed, the access to sufficient product-related information allows customers to detect weak points and built-in obsolescence. Low incentive is introduced thus, for companies to design and produce fragile, flimsy products.
•	H2 – Participative production supports longer service lives due to increased repairability. The users’ participation in the fabrication of their products creates a learning effect generating the necessary know how for them to repair those products. This is also reinforced by the accessibility of product related information allowing users building this know how even if they have not participated in its production by themselves.
•	H3 – Finally, the participation of the end-user in the design process may support better matching up of the product and the user’s needs. From an environmental perspective, better adaptation means avoidance of over-engineering and corresponding unnecessary environmental burden, as well as a closer emotional link between the user and the product, ultimately promoting longer product life to boot.
•	H4 – As participative production, at least in its today’s context of emergence, is motivated by an ideal of ‘do-ocracy’, where ‘makers’, in contrast to ‘consumers’, have access to decision-making power through their creative, constructive, proactive participation. From a very broad perspective, this ideal challenges the concept of social distinction through conspicuous consumption and even gives rise to a hypothesis of reduced consumption volume.
•	H5 – Participative production is performed in locally-bound value creation chains. This supports the use of local resources, promoting shorter transportation loops, adaptations to the local ecosystem (i.e. avoidance of unmanageable waste and non-exhaustion of resources) and even closed-loop material circles.

There is to date no empirical evidence validating those hypotheses. Drawing conclusions on the sustainability of participative production would require investigating complex and dynamic socio-economic effects, the consideration of which lies far beyond the scope of engineering science. Empiric approaches addressing the environmental efficiency of participative production focused so far on comparing additive and subtractive manufacturing (e.g. Kreiger and Pearce 2013; Faludi et al. 2015; Barros and Zwolinski 2016) and on environmental management practices in makerspaces (e.g. Prendeville et al. 2016). These studies delivered a contrasted picture, highlighting for example the fact that lower production scales imply lower efficiency per unit of service or that personal driven production may lead to abundance of gadgets.

Empirical study

The following section strives to bring empirical information into the debate by addressing this topic through a simple proxy question: How do existing open source hardware products aimed at participative production incorporate aspects of environmental friendliness? In other words, what are the environmentally friendly design features they implement? 18 existing open source products have been screened with the help of 11 eco-design aspects, in pursuit of insights on the environmentally-friendly design features used in current practice of OSPD. The first following subsection describes data acquisition and evaluation methods used to address this question. The corresponding results are presented in the second subsection.

Methodology

The methodological approach followed in order to address the above mentioned questions implied defining two criteria sets:

(1)	An analysis grid allowing highlighting environmentally friendly design features of selected products.
(2)	A criteria set for the selection of products to be screened using the above mentioned analysis grid.

Analysis grid

The integration of environmental-friendly features in the product designs was screened according to ten environmental criteria based on the ‘ten golden rules’ of eco-design developed by Lagerstedt and Luttropp (2006). This set of rules includes environmental aspects covering the whole product lifecycle. In addition to these criteria, it has been assessed whether the function the product fulfils is tied to an environmental advantage of some kind – that is, whether the product is supposed to replace a presumably less environmentally-friendly system. An example is the design of a cargo bicycle aimed at replacing car transport. In that vein, it is not only evaluated whether the products incorporate environmental friendly design features in the sense of the ten golden rules of eco-design, but also whether its development is motivated by sustainability ideals.

The resulting eleven criteria given in Table 1 have been evaluated for each product qualitatively based on secondary data, i.e. through the screening of the publicly available online documentation on the selected products. Criteria have been considered as binary: either the product presents one or more design features implementing the environmental aspect covered by the criterion, or it does not. For example: criterion 9 (recyclability/reusability) applies when a product is reportedly made of recyclable or biodegradable materials, criterion 3 (lightweight) when a product is supposedly made of lightweight carbon-fibre materials, and 10 (reversible joining elements) when the parts of a product are assembled with nuts and bolts. A criterion has been identified as not satisfied when no or insufficient information was found that allows interpreting the criteria as being satisfied.

Table 1. Analysis grid.

Beginning of life	1	Is the product made of recycled or reused raw materials and components?
	2	Is the product made of non-toxic raw material?
	3	Is the product designed to use little material? (lightweight design)
	4	Is the product designed to reduce transportation in the production and distribution phase?
Middle of life	5	Is the product designed to minimise energy and resource consumption during its use phase?
	6	Is the product designed to be repairable and upgradable?
	7	Is the product designed to be durable, e.g. is resistant to wear and damages and requires low maintenance?
	8	Is the product designed to be easily maintained?
End of life	9	Is the product designed to be recycled or reused at its end-of-life?
	10	Are components assembled with reversible joining elements so the product can be disassembled at end-of-life?
Product function	11	Is the function the product fulfils tied to an environmental advantage?

Note that the application of this analysis grid is not aimed at delivering a quantitative environmental assessment of the selected products in the sense of life cycle assessment. The objective is to reveal in a qualitative way the presence or the absence of environmentally friendly design features in the screened products.

Product selection

The study is limited to engineered mechanic or mechatronic products developed in an open source setting and to be produced in a participative production setting. Products have been selected out of a pool of open source hardware products previously gathered and published by Bonvoisin et al. (2016)⁷ according to following criteria:

•	The product is a discrete manufactured product, i.e. products of food and process industries have been excluded.
•	The product is first and foremost tangible and mechanical. It may include electronic hardware and consequently software. Purely electronic hardware or software products as well as non-solid physical products such as textile products have been excluded.
•	The product is rather complex in that it contains at least several parts. Products such as business card holders or cell phone cases made of one unique 3D-printed part do not fall into this category.
•	The product is developed for functional rather than aesthetic purposes. Jewellery, decorative items, gadgets such as personalised cell phone covers, or 3D printed rings do not fulfil this criterion and therefore are excluded.
•	The product is claimed to be open source by its surrounding development community and it satisfies the criteria of transparency defined by Balka, Raasch, and Herstatt (2014), i.e. blueprints and/or CAD files are publicly available.
•	The product is at least partly defined; undeveloped product concepts are not considered.

In addition to these criteria regarding product category and product openness, a supplementary criteria has been added here: the product is designed for participative production. The environmental performance of the products played no role in the selection or exclusion of the products.

Screening of selected products

Eighteen open source products were found to satisfy the selection criteria. References to these projects are given in Appendix 1. A detailed description of the qualitative analysis of four exemplary products is provided in the following paragraphs. Table 2 provides a summarised report of the evaluation of all considered 18 products.

Table 2. Evaluation of the 27 projects according to the 11 criteria.

LifeTrac is an open source tractor developed in the context of the project Open Source Ecology, a project aiming at developing and building a ‘Global Village Construction Set’, i.e. a set of 50 open source industrial machines allowing one to ‘build a small civilization with modern comforts’.⁸ Product information (CAD models, BOMs, manuals and videos) is available online and licensed under an extended Creative Commons license (CC-BY-SA 4.0 Attribution and ShareAlike). The product is designed to be ‘simple’ to assemble, following ‘DIY design’ principles, in order to allow ‘inexpensive manufacture’ in a local production context, hence addressing criterion No. 4 (reduced transportation). Simple design is achieved through the use of standard Lego-like parts that can be assembled by bolts. This design feature addresses the criterion 10 (reversible joining elements). The product is further designed to be modular in order to support product repair, maintenance and upgrade as well as module reuse, hence satisfying criteria 6 (repairability and upgradability), 8 (maintainability) and 9 (reusability). Modularity is achieved by ensuring disassembly through reversible connections (e.g. bolting of XYZ connections) and interchangeability of general purpose parts (i.e. components can be used in other products).

Multimachine⁹ is an open source multiple-purpose machine tool that is designed to be assembled by a layperson using commonly available tools from discarded vehicle parts. In its current documented version, it is a 3-in-1 machine providing the functions of drill press, lathe and milling machine. It is however intended that the product be designed as a platform for more tools. The project has been developed as a one-man activity and made open source via an online group. 80 pages of assembly instructions can be downloaded from the project website. No information about licensing could be found. The very concept of a product made of discarded car parts is supposed to ensure the sustainability of material procurement, hence satisfying the criterion 1 (reused content). The product is designed to be built in personal workshop based production setting, hence for local production, which satisfies the criterion 4 (reduced transportation). The concept of an extensible machine tool platform ensures upgradability and integration of new or updated machine tool functions using rotating elements (e.g. grinding) as well as reuse of parts. This satisfies criteria 6 (repairability and upgradability) and 9 (reusability).

WikiHouse¹⁰ is an open source house-building concept based on decentralised manufacture of wooden structural components. Structural components can be produced locally by CNC machines and assembled without specialised training or specific tools. Product information (e.g. CAD Models and building plans) is licensed under a CC-BY-SA license and is available online. A version of the product as a prepared kit is commercially available in the UK. The project claims to use sustainability-sourced timber and materials with low-embedded carbon which are recyclable and biodegradable. This addresses criteria 2 (nontoxic materials) and 9 (recyclability). It is designed for local production: the models can be downloaded and parts can be milled using local CNC machines and assembled on-site. This addresses the criterion 4 (reduced transportation). Energy in the use phase is supposed to be saved thanks to the high insulation capacity of the material used, hence addressing the criterion 5 (consumption in use).

RepRap¹¹ (for ‘Replicating Rapid-Prototyper’) is a general purpose ‘self-replicating’ desktop 3D printer initially implementing the process Fused Filament Fabrication (alternatively called Fused Deposition Modelling). The machine is claimed to be self-replicating as it can print some of the structural parts required for building a new machine. Several forks of this project have been developed since the release of the first functional machine in 2007. We consider here especially the model named ‘RepRapPro Mendel’. Intellectual property is licensed under GNU General Public Licence. Kits for building a RepRap are sold by several vendors. The machine is designed so structural parts can be locally produced and the whole product can be locally assembled as well. Local production is interpreted to satisfy the criterion 4 (reduced transportation). Parts are joined by nuts and bolts, the product can be disassembled and satisfies criteria 9 (reversible joining elements).

The criterion addressing the reduction of transportation in production and distribution phases (criterion 4) is addressed without surprise by all products (18 utterances), as they are all designed to be produced in a local production setting, i.e. at the point of use. Different strategies could be observed to achieve this: use of standard materials that are easily available for example in hardware stores, use of generally available discarded material, low constraints on mechanic tolerances in order to avoid high precision processing and therefore high machine investment costs, preference for processing steps involving commonly accessible tooling and avoidance of highly specialised tooling, preference for desktop machine tools such as 3D printers or low size carving machines, provision of prefabricated kits which can be sent in a condensed way and be assembled locally, use of ‘simple’ design.

The use of reversible joining elements is also a largely observed aspect (criterion 10, 15 utterances). This is the result of either a willingness to avoid complex joining processes like welding or to enable maintenance, upgradability as well as part reuse and separation for recycling. Most used joining techniques are nuts and bolts as well as Lego^TM-like interlocking parts using no additional joining elements.

The next most encountered criterion is the fact the function of the product is tied to a presumed environmental advantage (criterion 11, 8 utterances). Encountered functions are: production of local wind and photovoltaic energy, recycling of plastic, human-powered mobility, vegetable growing at the point of consumption, promotion of the repair of other products, and hosting bees.

Upgradability and repairability (criterion 6, 5 utterances) as well as reusability and recyclability (criterion 9, 4 utterances),are only partially encountered. These aspects are generally achieved through modular design as well as the use of reversible joining elements allowing for replacing defective modules along the use phase.

The least-addressed criteria are durability (criteria 7, no utterance), lightweight design (criteria 3, no utterance), energy use in the use phase (criteria 5, 1 utterance), and use of recycled materials (criteria 1, 2 utterances), and maintainability (criteria 8, 3 utterances).

Discussion

Thanks to local production, transportation of finished products may be avoided or replaced by the transportation of more bulky semi-finished products, leading thus to net savings. The use of local materials may even create incentives for closed-loop material circles. On the other hand, local and low-scale manufacturing entails the absence of economies of scale, more environmental impact per unit produced, and higher exposure to occupational health hazards. It can therefore be questioned how far low-scale manufacturing at the point of use represents an advantage in terms of sustainability. If the drawbacks outweigh the advantages, or in case this ratio isn’t readily apparent, then the exact parameters have to be identified by further research. Regarding the growing significance of citizen involvement in product manufacturing and of the local and low-scale manufacturing patterns illustrated in this article, it seems imperative to obtain a clearer overview of these decisive parameters in order to provide designers and citizens with appropriate guidance. In conjunction with that process, a proactive approach would involve questioning how far environmental impact can be prevented by the application of ‘design for local manufacturing’ design principles which allow extracting the most out of the environmental advantages of locally-bound production settings.

The function of most of the screened products bears a direct connection to environmental sustainability or is motivated by sustainability goals (e.g. cleaner energy or transportation, plastic recycling). This surprisingly high share may reveal an underlying correlation between environmental motivations and open source setting. Whether this correlation is contextual (the result of an ongoing trend) or structural, and how strong it is, may be the object of further research into the motivations and values of OSPD communities.

Finally, the present study underlines that modularity is being used by OSPD projects as a tool for achieving sustainable product design, especially through the use of simple and reversible joining elements along with a ‘construction kit’-like design. Modularity has been widely used for supporting extended service life of components and thus for reducing products’ environmental impact (Bonvoisin et al. 2016; Ma and Kremer 2016). Modularity supports ease of maintenance because it allows performing separate diagnoses and replacing damaged components. It furthermore supports the integration of new functionalities in the use phase and the reuse of components at the product’s end-of-life. At the same time, the empirical observation was made previously that product modular design plays an important role in OSPD projects because it supports the necessary self-selection of tasks required by a non-hierarchic organisation of work (Müller-Seitz and Reger 2010). An interesting area of future research would be to adapt the already large existing corpus of knowledge on modular product design to the specific case of OSPD, i.e. in order to identify how far product modularity can support the process of OSPD as well as the sustainability of the so-developed products.

The limited size of the study does not allow for drawing overall conclusions, but provides timely insights about the relationship between environmentally-friendly product design existing OSPD practices. In addition to the limited representativeness of the data-set gathered, two more biases associated with the use of secondary data may limit the validity of the results presented. First, the evaluation of the criteria is partly based upon claims made by the originators of the screened products. Obviously, just because it is said that a product is for example made of recycled material does not means that it actually is. Second, the online documentation describing the product may be incomplete. Just because no information has been found about how a product satisfies a defined criterion does not mean that this product does not satisfy this criterion. However, the reported screening allowed extracting interesting synergies between the fields of OSPD and sustainable product design which merit further research.

Conclusion

In this article, eco-design has been presented as a necessary though limited approach for addressing the challenges raised by sustainability in manufacturing. Complementary approaches such has Product-Service Systems have been introduced more recently to lead to more radical improvement in the sustainability of product and consumption patterns. This article introduced OSPD as an alternative proposition for the same objective. A definition and a characterisation of this concept have been introduced and put into the perspective of the current context of emergence of this new challenging though promising practice. The implications for sustainability of this concept have been discussed and formulated through five hypotheses, namely that it potentially leads to more robust design, increased repairability, better matching up of the product and the user’s needs, reduced overall consumption volume and production in locally-bound value creation chains. Based on the screening of the environmentally-friendly design features embedded in 18 open source hardware products, synergies between OSPD and sustainable product design has been looked for. This led to the identification of research questions for two identified synergies, namely participative production and product modularity.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Jérémy Bonvoisin, PhD, is a research fellow at the department of Industrial Information Technology, Institute of Machine Tools and Factory Management, Technical University of Berlin. He performed his doctoral studies on sustainable product development at the University of Grenoble with Prof. Daniel Brissaud. At TU Berlin, he focuses on open source hardware as a way to enable more sustainable production and consumption patterns. Along the precedent years of research, he covered research topics such as: remanufacturing, energy efficiency of electric and electronic products, environmental assessment of ICT-based services, energy efficiency of production systems and product modularization. Previous publications have appeared in the Journal of Cleaner Production, Journal of Engineering Design and International Journal of Computer Integrated Manufacturing. He has coedited the book “Sustainable Manufacturing – Challenges, Solutions and Implementation Perspectives” published by Springer.

Acknowledgments

The author is thankful to Kerstin Carola Schmidt and Mahdi Gheshlaghi for their participation in the data collection and evaluation process.

Notes

1. http://www.thingiverse.com/.

2. http://www.shapeways.com/.

3. http://reprap.org/.

4. https://www.apertus.org/axiom.

5. https://www.inventables.com/technologies/x-carve.

6. http://www.makerbot.com/.

7. An actualised list is maintained online (http://opensourcedesign.cc) and contains references of over hundred exemplary open source hardware products.

8. http://opensourceecology.org/.

9. http://opensourcemachine.org/.

10. http://www.wikihouse.cc/.

11. http://reprap.org/.

Appendix 1

Table

Name	Product category	Product	URL
RepRap Mendel	Machine tools	3D printer	http://reprap.org/wiki/Mendel
Wikihouse	Construction	House	http://www.wikihouse.cc/
OSE LifeTrac II	Agriculture	Tractor	http://opensourceecology.org/wiki/LifeTrac_II
Zoybar	Musical instruments	Guitar	http://www.zoybar.net/
Open structures’ CargoBike	Vehicles & parts	Cargo bicycle	http://openstructures.net/
Open source beehives	Agriculture	Beehive	http://opensourcebeehives.net/
Zoetrope	Power & electrical supplies	Wind turbine	http://www.applied-sciences.net/library/zoetrope.php
XYZ Spaceframe wehicles	Vehicles & parts	Bicycles	http://www.n55.dk/manuals/spaceframevehicles/spaceframevehicles.html
Multimachine	Machine tools	Modular machine tool	http://opensourcemachine.org/
Bloombot	Machine tools	Hydroponics system	http://bloombot.wikidot.com/
AeroSeeD micro-eolienne	Power & electrical supplies	Wind turbine	http://www.aeroseed.com/innovation/eolienne.php
Lyman Filament Extruder	Machine tools	Plastic recycler	http://www.thingiverse.com/thing:1199870
Hovalin	Musical instruments	Violin	http://www.hovalin.com/
Precious plastic	Machine tools	Plastic recycler	http://preciousplastic.com
Root washer	Agriculture	Root washer	http://farmhack.org/tools/root-washer
rBot	Machine tools	3D printer	https://rbotcnc.wordpress.com/
Aker kit	Agriculture	Gardening equipment	https://aker.me/
Kit zéro km	Clothing	Jeans	http://www.1083.fr/jeans.html