1. Introduction and Related Research
In tackling the various challenges identified in the transition towards a circular economy of more sustainable, intelligent, sociotechnical systems, various approaches have been proposed. These have varied from the 3Rs of Reduction, Reuse and Recycling of materials and energy [1
] to further versions with 6Rs and 9Rs frameworks [2
], which include remanufacture of materials as a key circular business approach. The concept of a “circular economy (CE)” is relatively new [4
], certainly in research—the first CE article for the Journal of Resources, Conservation and Recycling is recorded in 2007 [5
]. Remanufacturing, however, is not a new concept. According to Hatcher et al. [6
], remanufacturing has been a common industrial activity since the Second World War. As an academic research area, however, remanufacturing only emerged as a topic in the late 1970s and early 1980s. Robert Lund’s 1984 investigation [7
] into remanufacturing in the United States comprehensively communicated the experience of the US, as well as implications for developing countries. There has been a slow uptake of academic interest in remanufacturing, as compared to other circular approaches such as recycling [6
]. This may be attributed to the similarly slow uptake in industry interest in remanufacturing, particular in comparison to the recycling industry, which has grown by 300% [8
]. When thirty-seven (37) “design for remanufacture” papers were reviewed in Hatcher et al. (2011) [6
], it was found that 6 had been published in 1995–1999, rising to 12 between 2000 and 2005, and with only 19 published in 2006–2011, thus confirming the slow increase in remanufacturing-based research.
Within the literature, there are various definitions of remanufacturing, with [9
] and [10
] stressing that no universally accepted definition exists. Ijomah et al. [11
] state that this ambiguity has presented a key challenge for practitioners and researchers. In addition to the confusion in terminology associated with remanufacturing, differences in definitions have been seen to exist in the expected quality and performance level of a remanufactured product in comparison to a newly remanufactured product [12
]. Remanufacturing as [10
] “the process of transforming durable products that are worn, defective, or discarded to a ‘like new or better’ condition through a production-bath process of disassembly, cleaning, refurbishment and replacement of parts, reassembly, and testing”. Ijomah (2009) [13
] further argues that “the performance specification should be returned to the original level from the customers’ perspective and warranty will be given as equivalent to new products”. Chapman (2010) [9
], on the other hand, defines remanufacturing as “an industrial process of returning a used product to at least its original performance, equivalent to or better than that of the newly manufactured product”. The American National Standards Institute (ANSI), in a more recent definition, argued that “remanufacturing is a comprehensive and rigorous industrial process by which a previously sold, leased, used, worn, or non-functional product or part is returned to a “like-new” or “better-than-new” condition, from both a quality and performance perspective, through a controlled, reproducible and sustainable process” [14
]. Thus, remanufactured goods could be sold at a lower price, but potentially (or often) with a higher profit margin [15
]. Following these definitions, it can be observed that the focus on remanufacturing has grown from the state of the product, to the condition of the remanufactured product, to the warranty and to the broader subject of sustainability. Figure 1
gives an example of a used and a remanufactured end product.
Thus, remanufacturing makes extensive reuse of a product possible, hence keeping the product within a chain of circularity. This benefit is in addition to the demonstrated economic, social and environmental benefits of remanufacturing [6
] observed across key remanufacturing sectors. Research into remanufacturing in the UK by the Centre for Remanufacturing and Reuse in 2004 [9
] reveals that remanufacturing was found to take place in 22 sectors of manufacturing activity, of which the aerospace, automotive and mechanically powered machinery sectors were found to have the highest uptake. Polk (2013) [16
] states that 45% of gearboxes and 23% of engines in the aftermarket inventories of original equipment manufacturers (OEMs) are remanufactured worldwide.
Research related to remanufacturing is varied; however, it can be largely grouped into four categories; these include: (1) research into the processes in remanufacturing and design for remanufacture or DfRem; (2) the business models, frameworks and the wider supply chain associated with remanufacturing; (3) research into the benefits of remanufacturing; and (4) challenges experienced in remanufacturing.
Research on remanufacturing and remanufacturing systems has been conducted by several authors. This includes research on decision-making and Remanufacturing Decision-Making Frameworks (RDMF) for remanufacturing (Subramoniam et al. [17
], Ismail et al. [18
], Okorie et al. [19
]); remanufacturing production and lead times (Inderfurth and van der Laan (2001) [20
], Kiesmuller (2003) [21
]); design for remanufacturing (DfRem), (Hatcher et al. (2011) [6
], Ijomah (2007) [11
]); remanufacturing optimisation procedures for inventory level and Economic Order Quantity, (Kiesmuller and van der Laan (2001) [22
], Koh et al. (2002) [23
], Teunter (2001) [24
]); remanufacturing capacity planning, (Georgiadis, et al. (2006) [25
], Kleber R. (2006) [26
]) and research on assessing metrics for remanufacturability, Bras and Hammond (1996) [27
Within the scope of remanufacturing implementation benefits, several authors [15
] have identified specific environmental, social and economic benefits. These include a decrease in the use of resources, water and energy consumption, creation of employment opportunities, as remanufacturing is highly labour-intensive, as well as a 40–65% reduction in used material costs, and lower capital investment in factories and equipment acquisition [29
Technical Status of Fuel Cells
Since it was identified as an “environmentally aware energy supply” [30
], fuel cell (FC) technology has been used to replace energy supply systems such as batteries. According to literature [31
], the benefits of FCs revolve around their high efficiencies and low emissions. Hydrogen-powered fuel cells produce clean, pollution-free energy and have more than twice the efficiency of the conventional internal combustion (IC) engine [32
]. Comparatively, the traditional combustion-based power plant would generate at efficiencies of 33 to 35%, while fuel cell systems would generate electricity at efficiencies of up to 60% [33
] and higher with cogeneration [34
]. Sixty percent (60%) of the fuel’s energy is used in the fuel cell system. This corresponds to a more than 50% reduction in fuel consumption in comparison to a traditional vehicle with a gasoline IC engine. Composed of three active components, the fuel electrode (anode), an oxidant electrode (cathode), and an electrolyte in between them, a fuel cell is an electrochemical device that coverts the chemical energy of a fuel directly into electrical energy [34
]. The electrolyte is placed between the two electrodes, with bipolar plates on either side of the cell which help to distribute gases and serve as current collectors [33
]. The fuel cell provides an integrated cleaner alternative to the thermal processes involved in traditional combustion-based engines. Current combustion-based engines and energy generating technologies cause harm to the environment and contribute to many global issues, the most predominant of which is climate change [34
]. Fuel cells provide sustainable solutions and energy security in response to these issues. The static nature of fuel cells also means that they can operate without the challenge of noise or vibration [31
One type of fuel cell utilised in automotive vehicles is known as the proton exchange membrane fuel cell (PEMFC). This is widely regarded as the most promising for light-duty transportation [32
], as according to [35
], PEMFCs systems have a higher power density and lower operation temperature. In PEMFCs, a solid polymer is used as an electrolyte and porous carbon electrodes are also utilised as an electrolyte, [31
], as shown in Figure 2
. PEMFCs are normally fuelled with pure hydrogen supplied from storage tanks. The reaction in the fuel cell produces water heat and electricity as outputs [34
Air flows through channels to the cathode on the other side of the cell. When electricity has been produced (that is, work has been done), the electrons return and react with oxygen in the air and the protons (having already moved through the membrane) at the cathode. Water is hence formed, and this exothermic reaction [32
] generates heat, which can be used outside the FC.
The transportation industry is responsible for 17% of global greenhouse gas emissions every year [34
]. Thus, in the development of clean energy technologies, Electric Vehicles (EVs) have been touted as an excellent option for the reduction of emissions in the transportation sector [31
]. Currently, EVs are driven either by electrical energy stored in batteries or by fuel cell units. Power units in EVs are composed of two stacks, with each containing 40 fuel cells, making 80 FCs in all. There is a tank that supplies the hydrogen. Battery-driven EVs (BEVs) have the challenge of weight and recharging duration [36
]. This limits the driving range of traditional BEVs. Fuel cell electric vehicles (FCEVs) on the other hand, are superior to those with batteries in terms of volume, mass, initial greenhouse gas reduction and refuelling time [31
2. Remanufacturing and Industry 4.0
In the auto industry, for example, nearly all global players are producing electric vehicles (EVs) with parts which are sensor-embedded. This is expected to rise in the short term; many studies indicate that by 2020, over 11 million EVs will have been sold globally [38
]. Recycling has been proposed as the end-of-life (EOL) option for parts such as EV batteries (called rechargeable energy storage systems) and fuel cells, but there are many issues that make recycling less viable. The degraded battery that is taken out of the EV, for instance, still possesses around 80% remaining capacity [39
]. Recycling of such batteries reduces the active bulk of the batteries to material constituents, leading to the total loss of the remaining 80% of available capacity in the EV batteries [38
]. There have been concerns about the economic viability of recycling [38
]. For example, studies have questioned the ability of the market to absorb the enormous quantity of recycled materials, which may result in the long term in a situation where not all degraded batteries can be directly remanufactured into new batteries [40
]. There is also doubt regarding attaining the purity of recovered materials and sustainability objectives of EVs [38
Given the above issues specific to EV products, it is becoming important to seek other sustainable ways of extending the lifecycle of EV components. Remanufacturing is thus proposed primarily because it is seen as the most environmentally friendly of all of the EOL options [41
] and has the strongest tradition in the auto industry [38
]. As this research seeks to investigate the remanufacturing possibilities for the fuel cell, a sensor-embedded product, it is important to understanding remanufacturing within the larger context of Industry 4.0. Sensor-embedded products contains sensors that collect data from monitoring the product during its life-cycle [42
]. Product life-cycle data have an important positive influence on closed-loop product life-cycle management [42
]. OEMs, independent remanufacturers, third-party remanufacturers and maintenance teams all benefit from data from sensor-embedded products. These include receiving design feedback to improve upon the current design [43
], independent remanufacturers accessing the data conditions of used equipment for subsequent effective remanufacturing process planning [42
], and for OEMs with in-house remanufacturing units, their maintenance teams can schedule effective and exact maintenance tasks and spare parts scheduling using product life-cycle information from sensor-embedded products. The emergence of microelectromechanical systems (MEMS) technology has encouraged the manufacturing of smart sensors [42
], which are used in fuel cells and rechargeable energy storage systems (RESS) in electric and hybrid vehicles. According to [44
], basic smart sensors for product life-cycle data gathering will contain several essential elements. Following this, Ilgin and Gupta [45
] outline these embedded elements as: (a) a sensing device that registers environmental parameters, e.g., humidity, and produces a suitable readable signal form that can be analysed; (b) a microprocessor that processes the obtained signals; (c) a memory device which stores the received sensor data and the output from the microprocessor; (d) a data transmitter which transmits the data collected by the smart sensor to the communication network; (e) a battery or alternative power supply; and (f) an ID or sensor identification.
Industry 4.0 (I4.0) refers to the 4th stage of industrialisation, which aims for a high level of automation in manufacturing [46
] achieved via the adoption of ubiquitous information and communication technologies (ICTs). Coined and launched in Germany in 2011 [47
], and incorporated into its national “High-Tech Strategy 2020 Action Plan” [48
], I4.0 encompasses core technologies such as cyber-physical systems, cloud manufacturing, internet of things, and additive manufacturing [49
]. For current manufacturing processes, the emergence of Industry 4.0 portends revolutionary opportunities and enablement towards more sustainable manufacturing. As remanufacturing is generally labour-intensive and demanding of highly experienced personnel [46
], it is expected that remanufacturing of sensor-embedded products will face similar limitations. Other generic but significant challenges which affect an I4.0-remanufacturing integration include: low profit from remanufacturing [50
], lack of human resources, financial challenges [51
], legislation restrictions, lack of quality standards of remanufactured products, and customer perception [52
Despite the increasing interest and social, economic and environmental benefits associated with remanufacturing, challenges associated with remanufacturing exist. In a survey of 188 remanufacturing companies undertaken by the European Remanufacturing Network in 2015 [53
], several challenges faced by remanufacturing companies were underlined.
These included high labour costs, quality of feedback, lack of sales channels, lack of product knowledge, volume or availability, lack of technology and customer recognition. There is also a lack of knowledge in the assessment of remanufacturing technical and organisational processes [54
]. These challenges can contribute to, among other things, long and variable remanufacturing process lead times [55
]. Where large data sets are available, as in the case of fuel cells within the powertrain of hybrid and electric vehicles, it is becoming difficult to implement real-time and accurate remanufacturing on the shop-floor.
The remainder of the paper is organised as follows. Section 2
gives an overview of some terms and terminology utilised in the paper. In Section 3
, the materials and methods deployed in meeting the objectives of this paper are described. Section 4
contains the analysis and discussion. In Section 5
, the results are presented. The conclusions, limitations and considerations for further work form Section 6
and Section 7
4. Materials and Methods
In this section, the method utilised in identifying and itemising the data parameters required for real-time fuel cell remanufacture is presented.
4.1. Semi-Structured Interviews
To explore the data parameters for fuel-cell remanufacture, semi-structured interviews were undertaken. According to Robson (2002) [76
], this type of interview method “has predetermined questions, but the order can be modified based upon the interviewer’s perception of what seems most appropriate. Question wording can be changed and explanations given; particular questions which seem appropriate with a particular interviewee can be omitted, or additional ones included”. Hence, this kind of interview offers greater room for flexibility to the interviewer and interviewee and can be administered via a face-to-face interview session or via a phone/video conferencing call session. Semi-structured interviews are exploratory, using qualitative designs and can supporting quantitative research [77
In the choice of organisation and respondents to interview, an initial search for remanufacturing and automotive companies was undertaken using the “European Remanufacturing Network” database (www.remanufacturing.eu
). This database holds case studies of 66 companies across range of 10 key industry sectors. However, the focus was on the automotive sector, so an initial pruning was performed. Emails were sent and phone calls were made to the 19 companies who had automotive sector case studies recorded in the database. Out of these 19 companies, 9 responded, and further discussions were held. Three companies eventually agreed to participate. While this number is a fraction of the initial pool of 19 possible companies, Dyer and Williams (1991) [78
] argue that smaller case studies improve the capturing of greater detail regarding the context within which the problem studied exists. These selected respondents also showed a willingness to participate and share the deeper characteristics of their companies. The selected companies had ongoing collaborations with research Universities, and this was an enabling factor.
4.2. Case Companies’ Profiles
Organisational and respondent characteristics for this research are shown in Table 2
. Respondents are denoted by numbers (1)–(6). The companies selected for the study are innovative and leading companies in their respective fields of digital manufacturing research, remanufacturing, and automotive manufacturing. Below are the profiles for the participating companies.
Company A is a digital manufacturing and remanufacturing research company. It is an independent business, operating from three offices in the United Kingdom. It was formed in 2012 and is a strategic partner to the manufacturing sector, providing valuable services to government-funded programmes and private business as consultants and collaborators with academia. Their capabilities include advanced manufacturing research, virtual engineering, circular value chains, data analytics for resource efficiency, and manufacturing new technologies. Collaborators of Company A include Microcab Industries Ltd., MCT Reman Ltd., Env-Aqua Solutions Ltd. and Hydrogen London.
Company B is a UK and global leader in automotive remanufacturing and manufacturing. With over 45 years’ experience, they have served the automotive industry in OEM steering systems (manufacturing over 60,000 steering columns annually), military engineering, remanufactured steering, remanufactured hydraulics. The company is ISO 9001: 2015 certified and is a member of various remanufacturing and manufacturing networks across the world. Suppliers include JLR, Ford, Leyland Trucks, Arriva, Volvo, Textron, Caterpillar, and a host of others.
Company C is a UK-based hydrogen-powered fuel cell vehicle manufacturer. With a design studio in Spain, their business model is one which puts stakeholders, sustainability and profitability at the core of the business. Their prototype hydrogen fuel-cell car has a range of 300 miles on an 8.5 kW hydrogen fuel cell and with emissions of zero at the tailpipe (just water vapour). With an employee strength of 25, their partners include Michelin, KS composites, Sevcon, SDC Seat Design. Circular Economy is at the heart of what they do.
All companies asked to remain anonymous for this research. The respondents for the three companies were selected primarily because of their knowledge and experience in remanufacturing, sustainability and the fuel cell electric vehicles.
4.3. Data Collection
The researchers developed questions which requested respondents to give their views on the expanding parameters for data-driven remanufacturing. The data was gathered via face-to-face interviews; 2 were held at the companies’ on-site location, and the third, for Company A
, was administered via WebEx video conferencing. For data collection, a two-part questionnaire was developed. Part A consisted of questions relating to the company’s experience in remanufacturing. Part B consisted of questions on existing parameters for remanufacturing, new parameters for remanufacturing sensor-enabled components such as the fuel cell in electric vehicles, their rank of importance and efficiency calculation for fuel cell remanufacturing. The questions were developed based on an extensive review of literature and several discussions among the authors, and were first tested with the respondent from Company A
. This extensive evaluation of the questionnaire helped the researchers to produce a comprehensive list of questions and, based on the first feedback, produced a more objective list. After the first round of interviews, follow-up questions were sent to respondents via telephone calls and emails. This was done to ensure a thoroughness of the data collection process and to answer any criticism relating to issues of respondent numbers [12
Manufacturing companies implement various strategies in order to enable the transition to a more circular economy as well as enhance the performance and efficiency of the manufacturing systems. Among various circular strategies employed, remanufacturing, which restores used products to a like-new state, offers great opportunities to recover products and their parts while adding great benefits to the economy of the localities and countries where remanufacturing activities are carried out. In addition, remanufacturing requires less effort and resources for recovery, as well a retaining part of the raw materials and added value. This has made remanufacturing particularly profitable and viable for automotive-inclined companies, as evidenced in the research. With the entry and growing influence of electric and hybrid cars which are fuel cell- and/or battery-operated, it is becoming increasingly important to understand remanufacturing of the components of the electric cars within the context of Industry 4.0. This research was based on the hypothesis that data collected via sensors on the fuel cell can contribute towards their remanufacturing. The objectives of this research were hence set to identify and rank the remanufacturing parameters for the fuel cell, as well as to understand the relationships between data and remanufacturing using a simple CLD and SFD. Six respondents were selected from three remanufacturing/remanufacturing research companies identified as case studies.
It was found that variables required in the remanufacturing of the fuel cells are inextricably linked to the kind of data producing them. For purposes of remanufacturing, it is important to sub-categorise this data in two forms, namely: data from sensors and data from other sources. Within data from sensors, it was found that the vibration data which gives information about the physical state of the product was viewed as most important by the respondents within the context of remanufacturing. While the data from sensors are important, overall, for remanufacturing, the fuel cell is more dependent on data from other sources, named, “traditional variables”. For manufacturers this finding is important, as it would mean that there is need for greater collaboration between remanufacturers of diesel engine vehicles and electric/hybrid vehicles. Within the wider economic and environmental context, this provides evidence that data is important in enabling a circular economy.
Modelling results show that when components with information (data) increase, the inspection time for components decreases, and the remanufacturing cycle time also decreases. When the number of components with information increases, the number of components for remanufacturing also goes up. This puts pressure on existing capacity, encouraging management to reduce the components with information so as not to overload the system. Furthermore, from the simulation, it is seen that the system will not be able to cope if the components with information are increased without a corresponding rise in manufacturing. This is important for manufacturers in order to ensure sustained efficiency in the system.
Finally, it was observed that the values (built around the data) are more discretely occurring than dynamic. A part comes in for remanufacturing, the data is utilised and analysed, a decision is taken to remanufacture or not, and the process is completed. As a further work, it is recommended that other modelling and simulation techniques such as discrete event simulation are deployed in the investigation of the relationship between data from sensors and remanufacturing. This may uncover other values important to the remanufacturer. The modelling and simulation in this research was completed with hypothetical data which was approved by the team of experts constituting the respondents for this study.
Limitations and Further Research
For further research, it is recommended that results from sensor-embedded products, such as the FC or RESS, be utilised as a validation procedure where the data is collected over a longer period of time. Within the established cycle time, the BMS produces data for each of the identified parameters. It is recommended that real-time figures be extracted for the FC and other sensor-enabled battery components such as the rechargeable energy storage system as a means for validation. This could further inform useful relationships between data and remanufacturing.
One limitation in this model is that the results are essentially time-driven. Thus, important issues in the decision to remanufacture, such as cost, component value and remanufacturability of the component do not feature in the model. This could be an area for future research.
Rich and detailed answers were sought during the qualitative interview, with the list of questions prepared serving as a guide. However, while all of the respondents had remanufacturing engagements outside the UK, they were all based in the UK. This may be argued to lack representation, as opposed to if the respondents had been based across different global geographical locations. It is recommended that a wider sample of respondents be utilised in further research.
Information and data sharing by manufacturers and OEMs remains a challenge in research, as original data is increasingly viewed as a competitive advantage by companies. Thus, accessing data such as data from sensors for EOL research is a limitation of this research. While investigation of data sharing and collection paradigms is outside the scope of this research, OEMs and automotive companies can be encouraged to share data for research when there are viable benefits for them to do so.