Towards a novel Business, Environmental and Social Screening Tool for Product-Service Systems (BESST PSS) design

Product-ServiceSystems(PSS)haverecentlyregainedattentionintheliteratureandindustryduetotheirpoten-tial to contribute to sustainability. PSS are also enjoying renewed attention, currently, as the large societal focus on circular economy (CE) enforces the potential of PSS. However, PSS are not more sustainable than traditional offeringsbydefaultandthey,therefore,mustbedesignedwithsustainabilityinmindfromtheearlydesignstages andtheirsustainabilitypotentialscreenedalreadyataconceptualdesignstage.ExistingapproachestoscreenPSS sustainability have conspicuous shortcomings regarding their lack of comprehensiveness, usability and focus on theconceptual designstage deemed crucialfor thesustainability level ofthefutureoffering.There ispoor coverage, in the literature, of approaches to sustainability-driven PSS design, and no consolidated approach to supporting a comprehensive consideration of sustainability aspects in the early stages of PSS conceptualisation. This paper introduces a qualitative tool and a process that support decision-making through ex-ante screening of PSS concepts for manufacturing companies based on the triple-bottom-line (TBL) sustainable value potential over thePSSlifecycle. The tool was developed iteratively through three action research cyclesfocused on theory development andtheorytesting,withina manufacturingcompany.Thetooland theimplementation processare perceived as an effective and ef ﬁ cient way to screen PSS concepts through a comprehensive, yet readily applicable and usable approach by the industry practitioners. ©2022 TheAuthors.PublishedbyElsevierLtdon behalfofInstitutionofChemicalEngineers. Thisisanopenaccess article


Introduction
Never before has there been a stronger focus on both sustainability and servitisation in industry (Hallstedt et al., 2020).On one side, there is a rising trend among manufacturing companies to enhance revenue from through-life services (e.g.inspection, maintenance or repair) (Gebauer et al., 2010).On the other side, servitisation has the potential to decouple value creation from resource consumption, being a promising way forward for achieving sustainability (Kjaer et al., 2019).The implementation of service-based business models has the potential to extend the product life and reduce the need to manufacture new products, hence cutting costs and decreasing the environmental impact (Baines et al., 2007).
A service-based business model implies the development of more complex offerings, so-called product-service systems (PSS).PSS is a marketable combination of tangible products and intangible services (Goedkoop et al., 1999) that are life cycle oriented (Aurich et al., 2006) and supported by the infrastructure and the network of actors, designed to deliver more value than traditional transactional offerings (Mont, 2004).
PSS, a growing research field since the nineties (Tukker, 2015;Vandermerwe and Rada, 1988), is currently enjoying increased focus.The concept, nowadays popularly referred to as the "as-a-service" business model, has proliferated in recent literature, mostly due to advances in digital technology and circular economy (CE) (Pirola et al., 2020).Such business models are in close sync with recent policies in Europe, namely the Circular Economy Action Plan (European Commission, 2015) and the European Green Deal (European Commission, 2019) which emphasize their gravity in reaching the 2030 sustainability targets.CE aims to create the highest possible value and retain it for as long as possible while consuming fewer resources (Roy and Cheruvu, 2009), and PSS is seen as a possible key means of achieving that aim (Kjaer et al., 2018).PSS design is an integral part of servicerelated business development and there is a pressing call to support companies to integrate sustainability into their business development (Yang et al., 2017).Despite the intense development of tools and methods for PSS (Annarelli et al., 2016;Cavalieri and Pezzotta, 2012;Vezzoli et al., 2015), companies still face challenges when developing PSS (Pirola et al., 2020), especially in relation to sustainability (Qu et al., 2016).
PSS is not by default more resource-efficient nor sustainable than the sole product (Bech et al., 2019;Tukker, 2015), and may lead to magnified environmental impact or decline in social interactions under certain circumstances (Doualle et al., 2015).Hence the revived original focus on contributing to sustainable development (Brundtland, 1987) through PSS (Kristensen and Remmen, 2019) can be seen in newer PSS definitions (Annarelli et al., 2016;Vezzoli et al., 2015) that include the sustainability triple-bottom-line (TBL) (Elkington, 1998), which is the most common framework to implement sustainability in businesses (Palmer and Flanagan, 2016) and the academic literature (Purvis et al., 2019).PSS has the power to achieve a "triple-win" situation concerning the TBL; through economic aspects (e.g., cost and profit); environmental aspects (e.g., energy efficiency and product longevity); and social aspects (e.g., customer acceptance) (Chiu et al., 2018).
Since none of those benefits is guaranteed (Tukker, 2015), PSS offerings ought to be screened as early as the conceptual design phase to reinforce the prospect of developing sustainable offerings (Maxwell and Van der Vorst, 2003).For that reason, this research aims to create a tool to address the lack of consideration of TBL when screening PSS concepts in the literature (López et al., 2020), in the early stages of design, and to answer the following research question: -How to conduct sustainability screening of product-service system concepts in the early stages of design for the identification of triple-bottom-line benefit and cost hotspots?
To enhance clarity, the research question is further broken down into its specific elements: -PSS concept: an actionable (implementable) and assessable (screenable) design proposal describing the total PSS solution including the system's composition, its functionalities and business model that fulfil the requirements of the involved actor-network.-PSS concept screening: a simplified evaluation of alternative concepts to increase the likelihood of successful PSS development (Kim et al., 2013), which is more suited for the data-scarce early stage of design by its scope and level of detail.

Introductory literature review
The early-stage design is a crucial step in the development of sustainable PSS (Geum and Park, 2011;Sousa-Zomer and Miguel, 2017) and systematic early screening might avert resource allocation on design concepts with uncertain outlooks and risks of launching (Tran and Park, 2016).However, the early design phase is the most challenging phase, due to information scarcity and the inability to use dataintensive techniques (Rondini et al., 2020).Current methods have limited industrial applications; they are either too qualitative and unable to offer concrete concepts or too quantitative, which is expensive and time-intensive (Devanathan et al., 2010).
A common misconception in the PSS literature is that sustainability will be achieved if the designer follows the PSS design guidelines laid out, however, sustainability is not a given, when designing PSS (Tukker, 2015).Instead, a procedural sustainability screening is required to ensure that sustainability is designed into the PSS solution from the start (Maussang et al., 2009).Approaches to screen the sustainability of PSS are sparse in the literature, especially in the early design stages where it counts the most (Doualle et al., 2020).It is necessary to screen the sustainability of PSS as early as the conceptual design stage because most of the environmental, social and economic factors are determined in that stage (Bhamra et al., 2003), hence it is where truly effective impact can be made (Maxwell and Van der Vorst, 2003).
In the academic literature, PSS screening tools is one of the least addressed topics (Pirola et al., 2020).There is a significant concern regarding the scarcity of tools and processes to methodically screen alternative PSS concepts (Chen et al., 2015;Rondini et al., 2020), and to evaluate PSS performance both in terms of economic and environmental analyses (Annarelli et al., 2016;Tran and Park, 2016).Guiding principles to fully systematically screen the sustainability of a PSS are even scarcer in literature (López et al., 2020).Furthermore, there is a lack of support on how to design, develop and screen new value propositions, which are both more service-oriented and sustainable in all TBL dimensions (Qu et al., 2016).
Despite the current literature gaps, existing state-of-the-art tools and methods related to PSS screening were identified through a separate literature review and thoroughly analysed as a part of this research (Section 4.1), as explained in the Methodology.Furthermore, as presented in Section 4.2, an additional literature review was conducted to elaborate on the key dimensions to consider when screening PSS concepts.

Methodology
The research steps were carried out iteratively through a hypothetico-deductive approach (Gill and Johnson, 2002), by combining theory building and theory testing.The development and test of the proposed tool were carried out in the context of action research (AR) (Coughlan and Coghlan, 2002) in collaboration with a manufacturing company in the capital goods industry.
AR was deemed appropriate as it was a useful method to verify the early stages of research and test theories (Dyer and Wilkins, 1991).This particular company was selected because it is considered typical of many other manufacturing companies in the capital goods manufacturing industry (Sarancic et al., 2021).It is furthermore selected because the authors had a unique opportunity to observe and analyse the phenomenon previously rarely accessible to scientific investigation and interact directly and continuously with responsible decisionmakers as a part of the ongoing research project.
Despite the high level of practical relevance and the possibility to gain in-depth insights into the problem, there are certain threats to the validity of AR.Namely, the lack of impartiality of the researcher and the danger of being considered a consultant (Coughlan and Coghlan, 2002).These drawbacks are tackled by also utilising theoretical rather than solely empirical justification and with rigorous documentation in a cyclical fashion, unlike in linear consulting practices.
The research was structured in three cycles according to the four main steps of AR: diagnosing, planning action, taking action and evaluating action (Coughlan and Coghlan, 2002) (Fig. 1).AR is a participatory research method that involves taking action concurrently with knowledge creation and has been successfully used in PSS design (Tonelli et al., 2009).

Context and purpose
The empirical setting for the AR was a Danish mid-size machinery manufacturing company.At the time of the study (July 2021 to March 2022), the manufacturing company was in the early stages of PSS development as a means of contributing to the newly adopted TBL-oriented strategy.The company expressed the need to systematically select the PSS concept with the highest sustainability potential for further development, as they had very limited previous experience with PSS, and had only reactively and intuitively developed a few product-oriented PSS offerings.
A literature review of existing tools for PSS sustainability screening was conducted to identify potential tools to be used by the company.The search was conducted in the Scopus database by combining the synonyms of the keywords "product-service system", "sustainability" and "screening" and nine relevant tools were identified.The obtained tools were evaluated through collaborative quantitative data analysis (Miles and Huberman, 1994) together with the company representatives, resulting in the identification of a lack of existing tools suitable to support manufacturers in the early-stage sustainability screening of PSS concepts.

Cycle 1: theoretical development
1.1.Diagnosing.Empirical research, i.e. 10 hour-long semistructured exploratory interviews (Robson and McCartan, 2016) were conducted with seven case company managers.The interviewees were covering diverse functions (chief technology officer (CTO), aftersales, sales, corporate social responsibility (CSR) and sustainability managers) and seniority levels, ranging from three to 20 years of experience in managerial positions.The selection of interviewees was based on a strategic and purposive sampling strategy (Karlsson, 2008), covering diverse functions and seniority levels.The interviews aimed to identify the success criteria for the development of the PSS sustainability screening tool, informed by the previously conducted collaborative analysis of the existing tools in the literature.Therefore, the interviews revolved around the following question: "What success criteria should the tool for PSS concept screening satisfy?".The gathered data were consolidated, giving rise to the empirical success criteria for the PSS screening tool and eliciting the need to further explore the key dimensions for the screening of alternative PSS concepts (i.e., the TBL, life cycle thinking, value and PSS elements).
1.2.Planning action.An additional literature review to uncover the relation between PSS and the identified dimensions for PSS sustainability screening (step 1.1.)was conducted for theory building.Four separate searches were conducted in Scopus by combining the synonyms of keywords "product-service system" and the four dimensions.The gathered data were qualitatively analysed together with the company representatives to uncover the complex relation of PSS and the key dimensions identified for its screening.The analysis included the evaluation of the detail level that the decision support tool should be able to provide, concerning the elicited success criteria.This step also covered the planning of the development (1.3.) and the evaluation (1.4.) steps in this AR cycle.
1.3.Taking action.The first version of the tool was developed based on the combination of the review of existing tools, the empirically elicited success criteria, and the identified key dimensions through a series of workshops and meetings with the relevant stakeholders.The interactions focused on the development of the content and the logic of the tool while ensuring conformity with all the elicited success criteria and the key dimensions obtained (step 1.2).This step consisted of four workshops and two follow-up meetings with the company representatives.The first version of the decision-making tool was built iteratively by following the elicited success criteria, starting from the overarching strategic aspects and then progressing towards more operational aspects.
1.4.Evaluating action.The first version of the tool was tested through three two-hour-long workshops.The first workshop was focused on the evaluation of the logic of the tool and was supplemented by the second workshop (with two external industrial experts in service innovation).The third workshop (conducted between the authors and external experts) was focused on evaluating the tool's user interface and functionality.Improvement opportunities were identified both in the logic and user-friendliness of the tool through the application of the tool to different PSS concepts.A thorough re-alignment of the definition of the key dimensions and their inter-relation was the main outcome of this step.

Cycle 2: practical development
2.1.Diagnosing: N/A 2.2.Planning action.The planning of the second AR cycle included a series of meetings for the development of the second version of the tool, based on the identified improvement opportunities in the first cycle.Furthermore, this cycle involved the planning for the further test of the tool with master students, to enable broader applicability in different contexts.
2.3.Taking action.The second version of the tool was built based on the evaluation described in step 1.4.The contents and the logic of the tool were further detailed, leading to a consensus on the definition of the key dimensions.This was done through meetings between the authors, followed by a meeting with the company representatives.In this step, a visual representation of the concept screening tool was devised, as the case company found it crucial to be able to easily communicate and compare alternative concepts.Then, four distinct implementation processes were devised to make the implementation process simpler and quicker to use.

Evaluating action.
The second version of the tool was tested in a workshop with 17 students, in four randomly formed groups of 4-5 students, attending a PSS master-level course at the Technical University of Denmark.The evaluation focused on the tool's user interface and functionality.A dominant tool implementation process emerged in terms of the sequence of steps that enable the greatest simplicity and agility.To retrieve qualitative empirical data and ensure the validity of the student feedback, interviewing, direct participation and observation (Robson and McCartan, 2016) were used during the workshop.A feedback survey was conducted after the workshop in which students rated the feasibility, usability, and utility of the tool.3.3.Taking action.The third version of the tool was built to include the considerations step 2.4, i.e., the incorporation and detailing of the implementation process.The development was conducted through two meetings between the authors, followed by a meeting with the company representatives, where the implementation process was streamlined and presented to the company.The outcome of this step was a tool ready to be used with the case company.
3.4.Evaluating action.The third version of the tool was tested in the case company in two separate workshops, lasting 2 and 3 h, respectively.The workshops were facilitated by the authors and included previously interviewed managers from the case company.In these workshops, primarily success criteria related to the functionality of the tool were tested (i.e., whether the tool can support decision-making regarding concept screening and selection).

Context and purpose: review and analysis of tools and methods
Despite being one of the least addressed PSS-related topics (Pirola et al., 2020), nine approaches (Fig. 2) were identified in the literature that serve as a knowledge base for the development of the novel approach presented in this paper.None of these approaches explicitly answer the research question but each makes a specific contribution to the understanding of either PSS concept screening or PSS sustainability screening.The authors acknowledge numerous other approaches for PSS concept screening present in literature (see e.g.(Bertoni et al., 2018;Montelisciani et al., 2015;Sakao and Lindahl, 2012;Song et al., 2021)), however, these nine particular approaches were selected because they intentionally prioritise two or all three dimensions of sustainability in the most comprehensive way.The nine approaches identified in the literature tackle PSS screening from various perspectives, and they were thoroughly analysed to capture their contributions and shortcomings, thus paving the way for the development of an approach that is comprehensive, usable and focused on the conceptual design stage screening.
The existing screening approaches (Fig. 2) are disparate in their scope and the dimensions that they propose to be included, in the sustainability screening of a PSS concept.Due to the identified shortcomings in those approaches, none of the listed approaches was deemed appropriate for useon their ownfor PSS sustainability screening, as was also confirmed in the setting of the empirical case company, by the committee consisting of authors and case company representatives.Therefore, a PSS sustainability screening tool that would unify the most important contributions and overcome the shortcomings of the existing approaches (Fig. 2), with a clear focus on early-stage concepts and a straightforward implementation process, was deemed necessary.The main outcome of this cycle was the identification and consolidation of the success criteria for the development of the sustainability screening of PSS concepts (Fig. 3).

Step 1.2. Reviews and analyses of the key dimensions
In the following subsections, the findings from the literature reviews on key PSS dimensions (i.e., PSS elements, the TBL of sustainability, value and the PSS life cycle) are outlined.4.2.2.1.PSS elements.Despite the wealth of publications in the field, the PSS literature has not yet reached a standardised definition of the structural architecture of PSS, i.e., the constituent elements of the PSS concept that have to be designed to deliver the value proposition (Tukker, 2015).To be able to define the structure of a PSS concept, a literature review of PSS definitions was carried out to identify the PSS building blocks (Fig. 4).
The most recurrent elements in the PSS definitions are products, services, network and infrastructure.Although most authors agree that PSS can be split into products and services, fewer include network and infrastructure as core elements.Tonelli et al. (2009) argue that PSS is strongly context-related and to evaluate it means to analyse the internal and external ecosystems of a company.Therefore, collaboration with the involved stakeholders from the early stages of PSS design all the way through PSS operation is crucial for the development of steadfast and long-lasting customer relationships through PSS offerings (Fernandes et al., 2020).The infrastructure enables value delivery through supporting facilities, logistics, tools, soft and hard products, etc., which directly increase the value of the PSS provider's assets (Xing et al., 2013) and differentiate the provider from the competition.
This analysis was substantiated by examining the characteristics of PSS definitions with the highest citation index (Haase et al., 2017), where the most recurrent PSS characteristics were found to be product and service, customer needs, network and infrastructure.Furthermore, the most prevalent aspects of the PSS definitions by Annarelli et al. (2016) are customer needs, tangibility and intangibility, the systemic concept and networks and infrastructure.Therefore, once the customer needs are identified and the value proposition is formulated, the elements that mutually exclusively and collectively exhaustively describe the structure of a system that is PSS can be reduced to the most dominant definition, i.e., the seminal work of Mont (2004).In that definition, product, service, network and infrastructure, described by Pirola et al. ( 2020) "all" the PSS components, have to be intentionally designed to deliver value to the customer over its life cycle.However, PSS is more than just a collection of elements because

PSS sustainability.
A common assumption in the PSS field is that sustainability would be achieved if the designers follow the design guidelines, however, real screening is lacking in the early design process (Doualle et al., 2020).PSS is neither inherently sustainable nor circular, and it has to be purposely designed (Pigosso and McAloone, 2015).
Although PSS has the potential to achieve a "triple-win" concerning the TBL of sustainability, there are very few attempts to assess PSS from the TBL perspective in literature (Chiu et al., 2018;Lee et al., 2012).Observed strictly from the manufacturing perspective, different PSS archetypes (i.e., product-oriented, use-oriented, and result-oriented) have different potentials for the reduction of environmental impact (Tukker, 2004), which has to be considered when selecting and designing concepts.Product-oriented PSS implies the sales of products with add-on services (e.g., consulting or maintenance.Unlike the productoriented PSS, use-(e.g., renting, sharing and leasing) and resultoriented PSS (e.g., the focus is on the result without specifying the exact product) are distinguished through product ownership retention by the provider (Baines et al., 2007).
Sustainability is rarely treated as a primary root of value, it is rather seen as an add-on (Yang et al., 2017).Companies frequently comprehend sustainable value as the capacity to transform social and environmental value into business, but the social and environmental value should be seen not as enablers of economic value, but rather as standalone absolute values (Hart et al., 2003;Kristensen and Remmen, 2019).According to Bocken et al. (2015), a sustainable value proposition consists of the main dimensions: economic, social and environmental.Hart et al. (2003) define sustainable value as multidimensional "shareholder wealth that simultaneously drives us towards a more sustainable world".
A concept similar to sustainable value is the shared value (Porter et al., 2011) in which companies concurrently advance economic and social circumstances in the communities they operate.While (Baines et al. (2007) talk about the potential to decouple value creation from resource consumption through PSS, the concept of shared value (Porter et al., 2011) in which economic prosperity is coupled with social progressivity is often neglected in the PSS field (Merli et al., 2018).However, the significance of associating social issues (e.g., labour rights, social justice and communities) with sustainable PSS (e.g., social actors or stakeholder networks) is crucial to innovation and change (Chen, 2018).Both concepts share the same challenges, i.e., to decide which practices and actions to pursue and how to deliver the most value from the TBL perspective (Hart et al., 2003;Porter et al., 2011).
Table 1 Comments and the actions taken to address the feedback from step 1.4.

Feedback number and keywords
Action taken 1.Life cycle point of view Adoption of the provider's life cycle perspective, as defined by Aurich et al. (2006), places most of the customer's activities in the provider's MoL, excluding the purchase (BoL) and prospective decommissioning (EoU).

Life cycle boundaries
PSS life cycle is adopted as the operating frame (Sundin, 2009), which is not necessarily tied to that of a product, nor is it in sync with that of a service (Wiesner et al., 2015).E.g., a machine manufacturer can agree with a customer to provide a use-oriented PSS (Tukker, 2004) for a period of time.Within that period the manufacturer might be forced to replace multiple products, to respect the availability-based agreement.Therefore, within the same PSS life cycle, products or services can have their respective life cycles (BoLs, MoLs and EoUs) within the MoL of the PSS life cycle, which can be upgraded or downgraded due to ever-changing customer requirements within the life cycle of a PSS.

Four dimensions
The implementation process was made more user-friendly by introducing an extra prior step (Sheet A, Fig. 7) which made it easier to use the tool by gradually building up all four dimensions through the implementation process.

TBL interpretation
A nested or interdependent TBL view (Isil and Hernke, 2017) is adopted to screen PSS concepts.The implementation process was made more iterative with questions referring to previous sustainability dimensions (previous layer of the BESST cube) after each layer is filled out.4.2.2.3.PSS value.Over the years, the notion of value has become focal in the PSS field (Bertoni et al., 2017;Tran and Park, 2015).The value considerations are particularly important in the early stages of design, where value-driven design methodologies enable a more comprehensive awareness of the value of the total solution and help to avoid arriving at local optimum solutions (Bertoni et al., 2018).
As Chou et al. (2015) pointed out, to comprehensively assess the sustainable value of PSS, the connection between value and sustainability impact is a crucial factor to judge the sustainability performance of a PSS.Value should be used as a proxy for "PSS fitness" in the early-stage assessment, however, there is a lack of consensus on what aspects of value should be taken into account during the early-stage design assessment (Rondini et al., 2020).
A range of definitions of value exist, mainly based on the perceived benefits versus sacrifices (Toossi, 2011).Zeithaml's (1988) definition of value as the:" overall assessment of the utility of a product (or service) based on perception on what is received and what is given" remains the most prevalent definition.Value is a multidimensional concept that should not be reduced to only one form (monetary), or only one perspective (customer's), rather it should expand its focus beyond the limits of the company to include value for society and the environment (Kristensen and Remmen, 2019).
Many authors identify value in the same way as others define benefits, thus observing benefits and benefits missed (Yang et al., 2014), whereas according to the definition, value also consists of costs or sacrifices, therefore, value is created only when benefits surpass the costs (Figge and Hahn, 2004).This distinction is necessary also to avoid pursuing environmental or social benefits at any cost (Porter et al., 2011).Tao and Yu (2017) suggest that the integration of sustainability in the life cycle systems shows promise for sustainable value creation.Value has a perplexed and ever-changing temporal perspective, thus it also has to be observed from a life cycle perspective.Xing et al. (2013) put value into the life cycle perspective of a PSS by defining it as life cycle performance over life cycle burden.Furthermore, when comparing value delivery involving products versus PSS, the window to deliver value in PSS is much larger for all the involved stakeholders, in all three dimensions of sustainability (Yang et al., 2014).

PSS life cycle.
Life cycle thinking is a crucial concept for holistically developing sustainable PSS (Meier et al., 2010;Yang et al., 2014) and to guide decision-making for screening the PSS value over time (Xing et al., 2013).Life Cycle Assessment (LCA) and Life Cycle Costing (LCC) are often used to calculate environmental impacts and economic costs of PSS (Lindahl et al., 2014), however, the focus on the whole life cycle in the early design process is missing in literature (Pezzotta et al., 2018), although being crucial for successful value delivery (Wuest and Wellsandt, 2016).As a part of the complete product life cycle, the prolongation of the life cycle of the offering in the market is one of the main benefits of this holistic view because it: (i) increases resource efficiency; (ii) introduces more customer touchpoints creating longer business relationships; and (iii) allows for data accrual that might stimulate innovation and provide know-how feedback (Meier et al., 2010).
The basic product life cycle framework is most often described from the cradle-to-grave perspective in three phases: Beginning-of-Life (BoL), Middle-of-Life (MoL) and End-of-Life (EoL) (Corti et al., 2016).BoL encompasses design, procurement, production, assembly, and distribution.MoL includes the use, maintenance, repair, after-sales services, and insurance.EoL comprises reverse logistics processes including collection, disassembly, remanufacture, recycling and disposal.However, some authors argue that a more detailed description might be necessary for PSS (Wuest and Wellsandt, 2016): the PSS life cycle is a combination of a product and a service life cycle management that have to be integrated and coordinated (Wiesner et al., 2015).The two life cycles are often not aligned nor established in a closed loop to feedback information between them and BoL, MoL, and EoL (Corti et al., 2016).

Table 2
Comments and the actions taken to address the feedback from Step 2.4.

Feedback no.
Action taken

Life cycle transition
The exact transition point between BoL, MoL and EoU is blurred in PSS literature (Wuest and Wellsandt, 2016).Part of the confusion stems from authors dealing with consumer vs.capital goods.The distribution phase of manufactured goods is often prone to different classifications into BoL and MoL.In the case of capital goods (e.g., industrial equipment) the distribution phase falls into the BoL of a PSS, as such goods generate most of their value in use (Toossi, 2011) or through utilization, rather than ownership.Consumer goods discussions often place the distribution phase in the MoL of a PSS.For clarity in the tool implementation, the transition points between the three main phases are clearly defined in Fig. 5.

Repetitive
The implementation process is repetitive, but only for the first concept that is being screened.Once the next concept is screened, most of the cost and benefit drivers might remain unchanged in many alternative concepts.When put into the context of the PSS development duration, which may last for years in the case company's industry, the time required to fill out the BESST cube is relatively short.

Definitions
All the dimensions are clearly defined and presented in Fig. 5 which was then made available to the participants in the next AR cycle.

Prompts
The recommended implementation process steps were detailed and streamlined to include questions and prompts for easier workshop facilitation, as presented in Fig. 7.  2006) distinguish between the provider and customer life cycle perspectives, as they start at different points in time, which are the design for the manufacturer and the purchase for the customer.The customer is mostly focused on the usage phase, therefore, the provider should emphasize the same phase (Wuest and Wellsandt, 2016).
In summary, the key findings of the literature review carried out in this step are: (i) limited consensus on how to describe the main elements of PSS: product, service, network and infrastructure are selected as the main building blocks of a PSS concept, (ii) need to screen PSS concepts in the early design stages according to the TBL perspective, (iii) a value perspective is considered crucial when screening PSS concepts, and (iv) need to support the design and evaluation of PSS concepts from a life cycle perspective, with the beginning, middle and end of life stages.

Step 1.3. Development of the first version of the tool
The first version of the tool consisted of the four main dimensions: PSS elements, the TBL, value and the life cycle that had to satisfy the six elicited success criteria from the case company related to strategy, usefulness, scope, usability, communication, and perspective.
In this research, the TBL sustainable value of a PSS concept is defined as a function of cost and benefit in the three dimensions of sustainability (i.e., social, economic and environmental) over time, for each of the four PSS elements (i.e., product, service, infrastructure and ecosystem).The provider's perspective on the life cycle is adopted to screen PSS concepts, and it is split into BoL, MoL, and EoL.However, a more accurate term for replacing EoL would be the End-of-Use (EoU), which refers to instances in which the user can return a product at a point in the life cycle before its usefulness has perished (Östlin et al., 2009).

Step 1.4. Evaluation of the first version of the tool
The principal feedback and improvement opportunities identified related to (1) (as shown in Table 1) the lack of clarity on the life cycle point of view, (2) the indistinguishability between the product, service and PSS life cycle, (3) the ungraspability of the four dimensions at once and (4) the indistinct interpretation of the TBL.

Cycle 2: practical development
The second and the third AR cycles in the following sections include the descriptions of the development and evaluation steps (steps 2.3., 2.4., 3.3.and 3.4.explained in the Methodology).

Step 2.3. Development of the second version of the tool
The Business, Environmental and Social Screening Tool (BESST) for PSS concepts was proposed as a three-dimensional visual representation combined with the fourth value dimension (Fig. 5).The examination of  the PSS elements over its life cycle in the three sustainability dimensions provides a comprehensive picture of the benefits and costs of a given concept from the provider's perspective, supporting decision-making.
In this baseline, a PSS concept can be represented by 36 dice forming the cube, where each dice holds information about its value, i.e., benefit and cost, totalling 72 data points to screen a PSS concept.The division of a life cycle into cost elements was proposed by Schuh et al. (2009), who argued that a cost element determines costs incurred by resource consumption in a specific life cycle stage.This work expands the existing cost elements to cover all three pillars of sustainability as well as introducing benefit elements.Therefore, cost and benefit elements determine the economic, environmental and social cost and benefit incurred at each of the three stages of the life cycle for all four PSS elements.
The tool also utilizes the concept of vertical coherence (Joyce and Paquin, 2016) introduced in the triple-layered business model canvas, which supports the alignment of cost and benefits across the three sustainability layers of the BESST cube.
Table 1 describes the implementation of the improvement opportunities identified in step 1.4.

Step 2.4. Evaluation of the second version of the tool
Several improvement opportunities emerged from the workshops in the tool evaluation, where four different implementation processes were tested.The main improvement opportunities concerned (5) (as shown in Table 2) the transition point between the three PSS life cycle phases, (6) the implementation process repetitiveness, (7) the exact definitions of each dimension and (8) the lack of prompts to facilitate the implementation process.
The implementation processes in the workshop with students differed in terms of the starting points (i.e., the four dimensions of the BESST cube) and the sequence of the subsequent steps.
By analysing the data gathered through participant observation and the survey, the dominant implementation process was selected as the one used by Group 2, depicted in Fig. 6, which rated the feasibility and usability of their implementation process noticeably higher than the other groups.That implementation process is presented in Fig. 7, and its main difference from the other implementation processes is the introduction of the pre-step (Sheet A).  3 summarises the actions taken to address the improvement opportunities.

Case application
The BESST was used to screen a PSS concept brought forward by the case company.The screened concept was a use-oriented PSS where a food-processing machine is leased for a set period with a promised performance, but the customer is responsible for the function fulfilment.
Fig. 8 depicts the stage in the BESST implementation process where the hotspots have been identified for the given PSS concept in terms of benefit and cost drivers that most significantly influence this particular concept.The hotspots can be defined as concerns of significant impact potential (Bertoni, 2019).The crucial identified hotspots are depicted in Fig. 8 with two colours, green for the positive hotspots, and red for the negative hotspots, which are coloured in various dice of the BESST cube.
Arguably the most concerning negative hotspot for the case company is the monetary cost and related risks of the infrastructure in the BoL of PSS, for two reasons.First, due to the lack of capabilities to develop and integrate such infrastructure involving many different elements internally such as all the monitoring equipment, software to support automatic periodical billing, software to track and manage the service agreements over time, but also the more material-intensive infrastructure in terms of establishment of additional repair facilities and tools, as well as the expansion of the logistics infrastructure to cover the increased need for service.Second, due to the uncertainty concerning the scale of the future PSS market, i.e., the number of customers that the infrastructure must be able to support in the long run.Both challenges could be ameliorated by entering partnerships and sharing the responsibilities within the actor-network.Furthermore, the first challenge could be addressed either by internal capabilities development or the acquisition of external capabilities, while the latter  challenge could be addressed by thorough market analysis and customer interviews to estimate the size of the addressable market and attempt to project it in the future.
The positive hotspots are many in all three dimensions of sustainability and they add up to a potentially very beneficial business model for the case company.Most of the benefits manifest in the MoL of PSS which ought to become a prolonged period of product useful life and revenue stability from locked-in customers relieved from many of the service responsibilities and difficulties that were present in the pure product sale.

Discussion
The academic contribution of the novel tool manifests in eliciting and connecting the four crucial dimensions identified both from literature and empirically for sustainability screening of PSS concepts in the early stages of PSS design.This research contributes to a deeper understanding of sustainable value considerations in PSS design, and it enables industrial decision-makers to align the new business development to the company's strategy, which is based on the TBL.
The comparison of the BESST with the existing tools (presented in Section 4.1) with respect to the elicited success criteria (presented in Section 4.2) can be seen in Fig. 9.The comparison connects previously analysed existing tools in collaboration with the company representatives, the success criteria formed based on the analysis and the newly developed BESST which satisfied all the success criteria as judged by the company representatives and the authors.This comparison also serves as initial verification of the newly developed tool.
Elements of every tool examined were deemed useful for the application, however, most of the existing tools were considered too time-and resource-consuming to be applied in the case company by the authors and the company representatives.This is due to either lack of capabilities or lack of information in the company required to utilise the existing tools.Moreover, many of the tools lack the life cycle perspective, detailing to the level of the PSS architecture or a reflective screening perspective.
Numerous strengths of the BESST tool were discussed with the company representatives during its application.The tool was judged distinctly helpful to visualising and communicating the otherwise abstract phenomenon of PSS holistically, spurring many relevant discussions, especially about the eventual rebound effects, further refinement of the particular areas (dice in the BESST cube) of the PSS concept and even emergence of new concept ideas.
The applicability of the tool is judged to be plausible, even outside of the scope of its pre-determined success criteria, namely for ex-post screening and screening of wider business development initiatives, rather than PSS alone.Although the granularity level of the concept screening tool suffices the case company success criteria, the BESST tool's modular structure affords possible adjustments to accommodate for a more detailed screening, should the need arise (Fig. 10).
The four sheets utilised in the implementation process (Fig. 7) have been recognized as a helpful step to define the total cost and benefit of ownership (TCO and TBO) for any given concept, not only businessrelated but in all three dimensions of sustainability.
Practitioners with different backgrounds, knowledge and roles in the companies might have different perceptions of value, especially since strict indicators were not defined in each of the cells.A common example when using the tool is that people with commercial backgrounds overly focus on the business layer of the cube.Another common pattern is that the righter (towards the EoU layer) and lower dice (towards the social layer) are positioned in the BESST cube, the more obvious is the lack of competencies to identify and ameliorate the cost or capitalise on the benefit of the hotspot.A similar pattern can be observed with the infrastructure layer, which is more intuitive to identify the benefits and costs but lacks the methods to address the challenges arising in it.A possible reason for that observation is that the existing tools and practices do not focus as much on the PSS infrastructure, social and EoU aspects.The indicators were not defined to avoid constraining the thinking about the hotspots and allow for free ideation.To bridge this difference in perceptions, the recommendation is to include people with different backgrounds in the workshops while conducting the implementation process to ensure that the screening is considered from multiple angles.
Another important learning from the application of the final version is the importance of a completely clear definition of PSS concept alternatives before the concept screening with BESST.This is due to the possibility of the emergence of new sub-concept ideas during the screening process.This can result in an even better final concept but can also cause a new iteration of the initial concept development and prolong the whole process of PSS design.The limitations of the BESST tool are related to the static and fragmentary qualitative assessment of multidimensional PSS concepts.Thus, the concept screening as defined here is an aggregation of unconnected pieces that build up the whole PSS concept (Lee et al., 2012).Further limitations relate to the nature of AR which is conducted in a single case company.To ensure that the tool matures to general applicability across the capital goods manufacturing industry, the authors propose multiple case studies with a recommendation to conduct them both in smaller and larger companies than the one described here, and with different PSS archetypes (product-and result-oriented).

Conclusion
This paper introduced a decision-making tool for the ex-ante qualitative screening of PSS concepts with respect to TBL sustainable value along their life cycle.The BESST tool was built based on the literature and empirical findings through three AR cycles.The tool connects four dimensions identified in the literature as crucial for the sustainability screening of PSS concepts: PSS elements, PSS life cycle, PSS sustainability and PSS value and satisfies the case company success criteria elicited through the interviews.The industry practitioners involved in the study perceived the tool and the associated implementation process as an effective and efficient approach to screen the sustainability of PSS concepts.To address the tool's limitations, further work could focus on studying the connections and trade-offs between individual elements (dice) in the BESST cube.Other directions could tackle the tool's upgrade to contribute to the Circular Economy by focusing on Rstrategies, as PSS is seen as a possible key means to circularity.The BESST is to be further evaluated in other manufacturing companies, to ensure generalisability.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

3. 4 .
Cycle 3: industrial development 3.1.Diagnosis: N/A3.2.Planning action.The planning of the third AR cycle included the development of the third version of the tool and the planning of the evaluation workshops with the case company.

Fig. 2 .
Fig.2.Selected approaches that emerged from the literature review.

Fig. 3 .
Fig. 3. Case company success criteria for the tool.

Fig. 4 .
Fig. 4.An overview of proposed PSS definitions in the literature with highlighted PSS elements and depicted with pictograms (from left to right: product, service, network, and infrastructure).

Fig. 5 .
Fig. 5.The BESST cube for PSS concepts and the definitions of its dimensions.

Fig. 7 .
Fig. 7. Recommended implementation process.The BESST cube layers are sliced into three sheets (B, C and D) with a pre-step (Sheet A) to support the process.

Fig. 6 .
Fig. 6. Results of the student survey after the workshop.

4. 4 .
Cycle 3: industrial development 4.4.1.Step 3.3.Development of the third version of the toolTable2describes the actions taken to implement the improvement opportunities from step 2.4.4.4.2.Step 3.4.Evaluation of the third version of the toolImprovement opportunities identified in the evaluation with the case company included the need for (9) more examples of what the dimensions refer to in other companies, (10) a definition of the role of risks (potential costs) in the tool and (11) clarification of the duality of costs and benefits, as some of the spread across different dimensions with different impacts (e.g.monetary cost of training of technicians in the BoL brings social benefits in MoL)."Table

Fig. 8 .
Fig. 8. Examples of hotspot benefit and cost drivers identified through PSS concept screening in the case company.

Fig. 9 .
Fig. 9. Comparison of the BESST with other existing tools for PSS sustainability screening with respect to the elicited success criteria (SC).

Table 3
Actions taken to capture the improvement opportunities from step 3.4.