CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

Contents lists available at ScienceDirect

CIRP Journal of Manufacturing Science and Technology

journal homepage: www.elsevier.com/locate/cirpj

Towards a methodology to engineer industrial product-service

system – Evidence from power and automation industry
Giuditta Pezzotta a,*, Fabiana Pirola a, Alice Rondini a, Roberto Pinto a,
Mohamed-Zied Ouertani b
a
CELS – Research Group on Industrial Engineering, Logistics and Service Operations – Università degli Studi di Bergamo, viale Marconi 5, 24044 Dalmine,
BG, Italy
b
ABB Corporate Research Centre, Wallstadter Straße 59, 68526 Ladenburg, Germany

A R T I C L E I N F O A B S T R A C T

Article history: Manufacturing companies whose products have become increasingly commoditized are currently
Available online 19 May 2016 striving to identify innovative value propositions allowing them to re-position themselves in the market.
This is gradually leading to a business shift from delivering traditional transaction-based, product-
Keywords: centric offering to the provision of integrated product-service systems (PSS). However, the number of
Service Engineering companies failing in successfully pursue such a transition is still increasing. Consequently, Service
Methodology Engineering (SE), a discipline concerned with the systematic development and design of services and
Industrial application product-services, is gaining particular interest in both the scientific and practitioner communities. This
Simulation paper contributes to these fields by proposing a complete overview of the applicability of the SErvice
Product service system
Engineering Methodology (SEEM) in an industrial context. The SEEM aims at supporting companies
Case study
approaching the introduction of PSSs in their portfolio and suggests a structured decision-making
process to (i) define the PSS offering most aligned with company product(s) and customer needs, (ii) (re-
)engineer the (existing) service delivery processes, and (iii) balance the external performance (e.g.
customer satisfaction, delivery time, service cycle time) with the internal performance (i.e. efficiency) of
the service delivery process. The noteworthy benefits achievable through the SEEM are illustrated
through a real case at the industrial partner ABB – a multinational company providing power and
automation solutions. The implementation of all the SEEM steps is thoroughly described, and the
advantages experienced along with the difficulties encountered are highlighted. Managerial implica-
tions and the main gaps to address in future research are also discussed.
ß 2016 CIRP.

Introduction Therefore, these companies have to focus on services or service-

oriented products to succeed in the market. Thus, they need to
The recent economic crisis and the saturated and commoditized carry on with their traditional product design approach and to
market have led manufacturing companies to rethink their integrate it with proper service design as a mean to develop a
traditional business and move beyond simply providing goods [1]. marketable PSS. In addition, companies need suitable models,
These global trends, together with increasing market saturation methods and tools for collecting, engineering and embedding in a
[2], make the companies aware about the strategic relevance of the solution (bundle of product and service) all the knowledge that
provision of additional product-related services. This is perceived meets or exceeds people’s emotional needs and expectations
as a new source of value and competitive advantage by either [5,6]. Up to now, manufacturing companies have focused their
reactively fulfilling explicit customers’ requirements [3], or engineering capabilities on the pure physical product, neglecting
proactively providing them with new services or integrated the adoption of systematic engineering procedure for the
product-service systems (PSS) [4]. development of the service components of an integrated solution.
To this purpose, specific methods and models are required since,
even if provided in conjunction with a product, services are
characterized by high levels of intangibility, uncertainty and
* Corresponding author. Tel.: +39 035 2052005; fax: +39 035 2052077.
E-mail address: giuditta.pezzotta@unibg.it (G. Pezzotta).
simultaneity [2]. In this context, Service Engineering (SE) has

http://dx.doi.org/10.1016/j.cirpj.2016.04.006
1755-5817/ß 2016 CIRP.
20 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

emerged as a discipline calling for the design and the development process models, describing the steps needed to engineer a service,
of product-related service offering adding value to customers. and concrete methods, defining how to perform the model phases
In spite of the great success of the SE as an academic discipline, [21].
few of the methodologies available in literature can be directly As stated by Kimita et al. [22], several authors developed design
adopted by companies for two main reasons. Firstly, most of the methodologies for PSS under the Service engineering umbrella
methodologies identified are too complex or too many methods [13,23–26]. These researchers struggled with the definition of
are suggested (e.g. [7]). Secondly, the majority of them exclusively models and methods either to engineer the service component of a
focuses their attention on designing solutions able to satisfy PSS or to integrate the traditional product design and the service
technically customer needs [8–14]. In any case, they do not design through the development of a solution.
consider company internal performance. Therefore, balancing the By analyzing the most relevant works in PSS and SE fields [7–
external performance (e.g. customer satisfaction, delivery time, 14,27–33] two gaps have been identified: (i) they focus mainly on
service cycle time) with the internal performance (i.e. efficiency) customer perspective and (ii) they lack of critical and in depth
during the delivery of a product-related service has been evaluation of PSS performance in practice [15,34]. Recently, some
neglected in literature. To be sustainable in the long term, authors tried to overcome these gaps testing their methods in
companies need a methodology able to overcome the previous industrial setting [15,34–40]. However, they all have a strong
mentioned gaps [15]. customer orientation in relation to the design of the PSS service
The SErvice Engineering Methodology (SEEM) has been intro- components. Yoon et al. [41] consider both the customer and the
duced to fulfil this last challenge. SEEM proposes a set of methods company perspective but their work is limited to the PSS
that can be integrated with traditional product design and that can evaluation without considering its design. Also Pezzotta et al.
support companies in engineering and/or reengineering their [42] identify a way to consider both the customer and the company
offering. The SEEM structure supports in (i) identifying new perspective; however, the framework they proposed has not been
product-related service concepts fulfilling customers’ needs and validated in a real industrial environment.
(ii) identifying an efficient service delivery process balancing the To overcome the identified gaps, this paper proposes a
company external performance and internal performance. methodology validated at industrial level balancing the company
This paper aims at describing in detail the SEEM structure and at external performance with the long term business sustainability.
demonstrating its practical applicability through an implementa- For this purpose, a development process model and related
tion in a real industrial environment. concrete methods have been selected based on the literature
The paper is structured as follows: Service Engineering in the analysis on both SE and PSS design.
product service system context section presents a literature review Summarizing the most widespread models [7–13], four main
on Service Engineering with a focus on the models and methods common phases can be highlighted:
currently available. SErvice Engineering Methodology overview
(1) customer analysis: identification of customers’ features and
section describes the principal constructs of the methodology,
needs;
while SEEM on practice – industrial case at ABB section provides a
(2) requirements analysis: definition of product or service
full overview of the deployment of the methodology in a real
requirements addressing customers’ needs;
industrial environment. Discussion section summarizes the most
(3) PSS design: identification and design of solution(s) satisfying
relevant managerial implications while Conclusion section con-
customers;
cludes the paper and proposes further research prospects.
(4) PSS test and implementation: test the performance of the
identified solution and implement it.
Service Engineering in the product service system context

Designing and developing a PSS is a complex task due to the Concerning the methods, a wide range of authors has proposed
long and unpredictable lifecycle and the number of interactions alternative methods to carry out the above-mentioned phase.
among the actors involved and the constituting components [16– Table 1 lists these methods along with the phase where they have
18]. In fact, while in the area of product design a plethora of been adopted.
methods is widely accepted by the research community, in the area Most of the PSS design and SE literature highlights the relevance
of pure service and product-related service design such robust and of deeply analyze the customer explicit or latent needs. However,
common approaches are not available. Consequently, when only few methodologies in these fields clearly state how to collect,
compared to physical products, services are generally under- analyze and summarize those data. As reported in Table 1, the
designed and inefficiently developed [19]. The need of methods in Persona Model results as the most adopted method due to its
the area of service design is increasingly being recognized as ability to summarize in a visual way the data collected thought
relevant by designers, engineers and managers to create a market segmentation surveys or interviews.
successful solution, even though the knowledge on how to develop The identification of the PSS that can really answer to customer
a service and who should design it is still marginal [20]. This is the needs is carried out in the second phase. Among the different
main motivation behind the continuous growth of Service methods, most of the methodologies adopts Quality Function
Engineering (SE) as a technical discipline. Based on the definitions Deployment (QFD), since it represents a structured approach to
provided by Bullinger et al. [21] and Shimomura and Tomiyama define customer needs and translate them into product-service
[13], SE can be termed as a technical discipline concerned with the functions [49], and Functional Analysis/FAST – Function Analysis
systematic development and design of services, aiming at System Technique, which allows to translate the functions
increasing the value of physical artefacts. It is a rational and expected by the customer into functionalities and the technical
heuristic approach based upon the discussion of alternatives, goals, solutions [14].
constraints and procedures, through the adoption of modelling and The third phase deals with the PSS design. Here, service
prototyping methods. Accordingly, the aim of SE is to increase the blueprinting is the most used method since it allows representing
value of service offering by improving the service conception, the service delivery process from the customer perspective
service delivery and service consumption through the adoption of highlighting the physical elements that can be perceived by the
proper engineering methodologies. The development of a Service customer, and the activities where customer gets in touch with the
Engineering methodology implies the definition of development service provider [58].
 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32 21

Table 1
Summary of available PSS design and SE methods.

Phase Methods References

Customer analysis  Persona model [43–46]

 Cost–benefit analysis [47]
Requirements generation and analysis  Quality function deployment (QFD) [17,37,48–50]
 Benchmarking [17]
 Functional analysis/FAST – Function analysis system technique [37,45,50,51]
 AHP [50,52]
 Agent based simulation [53]
PSS design  Functional analysis/Function analysis system technique (FAST) [14,54,55]
 Service blueprinting [31,46,50,56–59]
PSS evaluation and implementation  Simulation (discrete event simulation and continuous simulation) [38,39,60–63]
 QFD [64]
 3D visualization [65]
 Failure mode and effects analysis (FMEA) [66,67]
 ANP [68,69]
 AHP [70]

Lastly, in the PSS evaluation and implementation phase, the application in several industrial cases. These applications have
designed PSS is assessed and, if satisfactory, implemented. Despite been carried out in collaboration with ABB, a leading company in
the several methods suggested, the majority of methodologies in power and automation technology. Continuous interaction and
literature adopts simulation (both discrete event and continuous close collaboration with ABB scientists and service managers
simulation) since it allows for the dynamic analysis of a system allowed to refine the methodology in terms of theoretical concepts,
under different and future conditions and scenarios [63]. methods and terminology making it more appropriate for the
Starting from the models and the methods used in literature, in industrial environment.
the next section the Service Engineering methodology (SEEM), Hereafter the SEEM steps and methods are described.
developed in close collaboration with practitioners and involving The SEEM, represented in Fig. 1, is composed of two main areas:
industrial feedback since the beginning, is proposed. The main the customer area and the company area. The former addresses
purpose of SEEM is to support companies in engineering and re- customer analysis while the latter aims at supporting the
engineering their PSS offering, taking into consideration both definition of a service delivering process considering both the
company internal performance and customer needs. company external and internal performance.
More in detail, the SEEM encompasses the most common phases
SErvice Engineering Methodology overview in the SE models, namely: offering identification and analysis,
customer needs analysis, process prototyping, and process valida-
As emerged in the literature review, one of the main gaps in the tion. As shown in Fig. 1, the first two phases belong to the customer
PSS design and SE is the absence of an industrial tested area, while the remaining two belong to the company area. In
methodology focusing on both customer perspective and com- addition, some of these phases are further decomposed into tasks
pany’s internal performance. This rather myopic view can lead to and, for each of them, one or more methods have been suggested.
the development of services fulfilling customer needs completely, In the remainder of this section, an overview of the four phases
but that can potentially undermine the company economic is provided. Moreover, since industrial companies often need to re-
sustainability in the long term, or vice versa. engineer its offering SEEM has been created in order to be
To achieve this goal and to ensure its industrial applicability, applicable also in this situation. Obviously, when it is applied to
SEEM has been developed starting from the theoretical background PSS re-engineering, the analysis starts from the comparison of the
presented in previous section and has been refined adopting an company offering and the customer needs in order to identify
iterative cycle of feedbacks and reviews collected during the existing or potential gaps.

Fig. 1. The SErvice Engineering Methodology (SEEM).

22 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

Customer area Requirement Tree (SRT) and a Quality Function Deployment (QFD)
based analysis are proposed.
The first two phases in (re)-engineering a PSS is the analysis of The SRT, is an ‘‘ad-hoc’’ method developed in the SEEM and it is
the customers’ needs and of the current solution portfolio (if any). based on the functional design domain knowledge. It is mainly
The aim is to identify the customer needs to be fulfilled by the PSS. drawn on the ‘‘Customer-Oriented FAST model’’ [54] and the ‘‘View
Model’’ [71]. However, the concept of ‘‘function/functionality’’
Offering identification and analysis behind these methods revealed complex to be understood by
This phase of the SEEM refers to the analysis of the current company representatives. Thinking about ‘‘what customer would
offering of the company and/or generally of the market. The aim is like from the company’’ resulted more intuitive and approachable.
to have a clear understanding of how the company is actually Thus, the SRT has been developed as shown in Fig. 2.
satisfying the customer needs.
a. Needs (N): needs express the customer necessities, in terms of
Customer needs analysis results of the expected PSS and/or performance. These needs are
The purpose is to obtain a clear understanding of the customers’ identified in the first phase of the SEEM, through the customer
needs and requirements in terms of products, service, and analysis. An example can be ‘‘maximize the plant availability’’.
expected performance. This analysis can also lead to the b. Wishes (W): they express how the customer wants to satisfy his/
segmentation of customers in several, homogeneous classes in her needs. For example, considering the need ‘‘maximize plant
terms of main requirements and needs [44]. Even if a specific availability’’, the wishes can be ‘‘reduce breakdown time’’ and
method to perform this analysis is not suggested, this step can be ‘‘increase plant lifecycle’’.
implemented in several ways, such as through market research, c. Design Requirements (DR): they represent how the company can
customers’ interview, focus groups or expert panels. satisfy customer’s wishes. In other words, they represent the
PSS(s) the company can offer to achieve customers’ wishes and,
Company area to fulfil accordingly their needs. For example, to fulfil the
customer wish ‘‘reduce breakdown time’’, possible design
Starting from the customer needs identified in the customer requirements can be ‘‘preventive maintenance’’ or ‘‘remote
area, the PSSs able to satisfy such need(s) are identified and the monitoring based maintenance’’.
product and service elements defined. Moreover, the service d. Activities (As) and Resources (Rs): they represent how the
delivery process(es) are prototyped, validated and added to the company provides a specific design requirement to customers.
company offering. Its aim is to give explicit information to the process design in
terms of main service delivery process activities (As) and
Process prototyping resources that can be both product components and human
The first phase of the company area is the process prototyping, resources (Rs). In case of product components, their design is left
which aims at identifying one or more PSSs and at designing the to traditional product design methods used in the company
associated service delivery process(es). This phase is further while for service-related activities and resources they are
decomposed in two tasks. discussed in the next steps of the methodology.

Requirements design. This task puts into evidence the main These four levels are hierarchically related in the SRT, as
relationships between customers’ need(s), PSS offering and the depicted in Fig. 2. Considering the literature findings, a method
resources needed to deliver the PSS. For this purpose, the Service based on the Quality Function Deployment (QFD) approach has

Fig. 2. Service Requirement Tree (SRT).

 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32 23

been adopted to identify the most relevant elements (DRs, the kth design requirement in satisfying the jth wish, while the
activities and resources) in fulfilling customer wishes and needs. weights WIj are those obtained at the previous step, as reported
Besides being the most widespread method in literature, QFD has in Table 3.
been selected because the way in which weights are set and how
the assessment of the tree elements is carried out, are readily The design requirement importance is calculated as follow:
understandable by industrial people. X
The QFD-based approach developed in the SEEM is articulated DRIk ¼ WIj Ikj 8 k 2 DR (3)
j2W
in four main steps. A weight is assigned by the people involved in
the PSS design (e.g. customers, service managers, sales people) to
each branch of the SRT in order to quantify the importance of the DRk
DRI%k ¼ P 8 k 2 DR (4)
lower level element in satisfying the linked upper level element. DRk̂
k̂ 2 DR
The possible weights are:
The QFD based matrixes linking design requirements DR –
activities A, and activities A – resources R, are based on exactly the
 ‘‘9’’: the lower level element is fundamental to fulfil the upper
same logic and allows the calculation of activities (AI and AI%) and
level element
resources (RI and RI%) relevance.
 ‘‘3’’: the lower level element is important but not fundamental to
The QFD based analysis supports the decision-making process
fulfil the upper level element
since it displays important information in terms of relative
 ‘‘1’’: the lower level element is not essential to fulfil the upper
importance among design requirements, resources and process
level element
activities in relation to the main customer need(s). In other words,
QFD helps define the activities and resources the company should
For each step, assuming a set N of needs and a set W of wishes
focus on to properly fulfil customer needs.
with cardinality h and v, respectively, a matrix is built a follows:
In conclusion, the SRT allows to identify new or updated PSS(s)
highlighting the service delivery process activities and resources,
 The first step evaluates the relation between needs i 2 N and
and the enabling product components. The SEEM does not tackle
wishes j 2 W, by measuring to what extent each wish is
the design and engineering of the product components, identified
important to satisfy the need. Such value is represented by a
as resources in the last level of the SRT, but it relies on traditional
weight Iji that can assume the values 1, 3 and 9 as previously
and widespread product design methods and tools.
illustrated. For each wish, an importance measure WIj is defined
and expressed as:
X Process design. This task involves the definition and representation
WIj ¼ NIi Iji 8 j 2 W (1) of alternative service delivery processes of one or more design
i2N requirement(s). As previously stated, this phase of SEEM, as well as
the following ones, focuses only on engineering (or re-engineering)
where NIi is the weight assigned to ith need. Similarly, the
the service component of a PSS (i.e. the service delivery process).
importance can be expressed as a percentage value as follows:
In the SEEM, the Service Blueprinting [57–59] technique is used
WIj for simultaneously depicting the service delivery process, the
WI%j ¼ P 8j2W (2)
ĵ 2 W
WIĵ points of customer contact, and the physical evidence of the service
delivery from the customer’s point of view. In particular, the
Table 2 shows the QFD based matrix linking needs and activities composing the process are classified into four categories:
wishes. (i) customer’s activities (performed by the customer), (ii) front-end
 The second step analyses the relation between the set of wishes activities (performed by the company interacting with the
W and the set of design requirements DR of cardinality d, customer), (iii) back-end activities (performed by the company,
highlighting what the company should provide to reach the but hidden from customer view), and (iv) support activities
customer wishes. Also here, a weight Ikj is assigned to each (general management activities performed by the company to
couple j 2 W and k 2 DR, expressing the relative importance of support several processes). In case of re-engineering, this phase

Table 2
QFD based matrix: needs vs. wishes.

Wishes Needs W importance W importance %

Need 1 Need 2 ... Need h

Wish 1 I11 I1h WI1 WI1

Wish 2 WI2 WI2
... Iji ... ...
Wish v Iv1 Ivh WIv WIv
Needs weights NI1 NI2 ... NIh

Table 3
QFD based matrix: wishes vs. DRs.

Design requirements Wishes DR importance DR importance %

Wish 1 Wish 2 ... Wish v

DR 1 I11 I1v DRI1 DRI1

DR 2 DRI2 DRI2
... Ikj ... ...
DR d Id1 Idv DRId DRId
Wishes weights WI1 WI2 ... WIv
24 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

entails the mapping of the current service delivery process(es) (as- 1. Company internal measures that can be set considering the
is) as well as the identification of possible alternative and improved company’s strategy and goals. The typical indicators belonging
delivery process(es). to this category are activities’ duration, waiting times, resources
Furthermore, in the blueprinting model the parts of the process utilization, costs and so. Usually, these indicators can be
corresponding to the activities identified through the SRT are measured directly through the simulation.
highlighted. This link allows for the identification of the most 2. Customer satisfaction based on cycle time. This indicator takes
relevant performance to be monitored in order to answer to into account the total service cycle time that is defined as the
customer needs. In fact, internal and external performance are total time elapsed from when a customer expresses a need to
measured at both single activity level and the entire service when that need is satisfied [76]. Service cycle time indicator is
delivery process level. obtained through the simulation results. The main concept
behind this indicator is that the lower is the cycle time the
Process validation higher is the customer satisfaction in relation to the selected
The result of the previous phase consists in the definition of one service [77,78]. Furthermore, the higher is the relevance of the
or more alternative service delivery processes. Nonetheless, this is a activity in satisfying customer needs (emerging from the QFD-
static result: nothing can be inferred about the performance of the based approach), the higher is the benefit of having low cycle
process(es) from internal and external point of view. Therefore, the time. For this reason, the indicator considers the cycle time
aim of the third phase of the SEEM involves the validation and with respect to the activity and the importance of the activity
assessment of the prototyped process(es), as well as the identifica- in satisfying the need (Ihk).
tion of the most suitable configuration of the process activities and
resources. To this end, the SEEM adopts a process simulation
approach, since it allows for the dynamic analysis of a system (the Therefore, the customer satisfaction based on cycle time related
service delivery process in this case) under different conditions and to each activity h 2 A and to each design requirement k 2 DR(SAhk)
scenarios. Considering that service delivery processes are fairly well has been defined as follows:
defined discrete processes [38,39,72–74], the methodology suggests
the adoption of Discrete Event Simulation (DES). DES has a great Ttarget A ðhÞ Ihk
SAhk ¼ P 8 h 2 A; k 2 DR (5)
potential as a means of describing, analyzing, and optimizing service T A ðhÞ I
ĥ ĥk
delivery processes [75] and supports their systematic and optimized
engineer. DES can be run with a wide range of software available in where
the market [61]. TA(h) is the average duration of activity h 2 A
The purpose of the simulation is to: (i) assess the performance Ttarget A(h) is the target duration to carry out the activity
of a service delivery process under different conditions (what-if h 2 A. Target duration can be either fixed by the company or
analysis), (ii) evaluate the effectiveness of possible process fixed as the minimum time obtained during the simulation of
changes, (iii) support the selection of the process configuration the as-is process configuration. In both cases the Ttarget A (h) is
with the best trade-off between internal and external performance, not the lowest possible, however it can be considered as a good
and (iv) provide insights about the service delivery process target.
dynamics and bottlenecks. Ihk is the importance of the activity h with respect to the design
The following table reports the validation procedure, highlight- requirement k as indicated in the QFD structure.
ing the difference between the engineering and the re-engineering Fixing the values Ihk, the lower is the ratio between TtargetA(h)
case (Table 4). and TA(h), the lower is SAhk, indicating a lower customer
Considering the main purpose of the SEEM, related to the satisfaction based on cycle time for the activity h on the design
identification of a proper balance between external performance requirement k. When TA(h) ! 0, SAhk is closed to the target value
(Customer satisfaction based on cycle time) and internal company and indicates a higher customer satisfaction based on cycle time.
performance (Company internal measures), two categories of KPIs The overall customer satisfaction based on cycle time indicator,
are assessed in the validation phase. They are: S, is calculated as the sum of the SAhk related to all the activities and

Table 4
Validation procedure.

Engineering case Re-engineering case

Check the solidity of the Once the process is simulated, the performance Once the process is simulated, the performance obtained (in
prototyped process obtained (in terms of number of jobs performed, lead- terms of number of jobs performed, lead-time. . .) are compared
time. . .) are compared with the desired with the actual performance. The simulation model is refined
performance. The simulation model is refined until it is until it is possible to assert that the model fits the industrial
possible to reality
assert that the model is realistic and fits the desirable
performance
Define the as-is future n.a. The as-is service delivery process model, refined in the previous
target scenario step, is simulated setting future company conditions (such as
forecasted service demand, updated service portfolio, and so
on) in order to understand how the actual company service
delivery process would perform under the forecasted changes
(as-is future target scenario)
Perform the Alternative service delivery processes are created, Starting from the as-is future target scenario, alternative service
what-if analysis simulated and compared delivery processes are identified, simulated and compared in
in order to define the best process configuration, namely order to define the best process configuration, namely the one
the one maximizing maximizing the trade-off between internal and external
the trade-off between internal and external performance
performance
 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32 25

design requirements previously identified: to avoid misunderstanding and to triangulate the data collected.
Those meetings have been also useful to keep the company
X X
S¼ SAhk DRI%k (6) involved in the research, collect feedback, update and modify the
k 2 DR h 2 Gk methods adopted, and early transfer the results to the business
unit.
where Gk is the set of activities that are involved in the design
requirement k (e.g. with reference to the SRT in Fig. 2, it is possible Customer area: offering identification and analysis and customer
to write GDR1 = {A1, A2, A3}). needs analysis
By construction, when S ! 0 customers are less satisfied with As previously mentioned, the methodology started with the
relation the service cycle time, whereas when S ! 1 customers result analysis of the current service portfolio. Currently, the 90% of the
more satisfied. When the duration of each activity is equal to the target total service revenues of the business unit is derived from the
one, S is equal to 1. S can result higher than 1, in the case DRk duration is following offering:
lower than the target one. This last situation is not common because
the target value is a good goal to achieve by definition.  Preventive and corrective maintenance. It refers to all the activities
Thus, considering the company internal measures and the performed on the customer’s product in order to make it functioning
customer satisfaction based on cycle time indicators, simulation is as efficiently as possible or, in case of corrective, to recondition it to
used as a decision making tool to test different, alternative proper functioning. In the analyzed case, maintenance can be
scenarios and process configurations, as well as to identify the best performed at the customer site or at the ABB plant.
one according to the pre-specified key performance measures.  Replacement. It consists in the provision of products currently out
In the next section, the implementation of the methodology in of production. A limited amount of these products is still
an industrial case of PSS re-engineering is presented and discussed. produced for customer with plant’s specific needs.
 Retrofit. ABB provides specific kits to adapt a new product to the
SEEM on practice – industrial case at ABB fixed part of an old one. This helps in adding functionalities to an
aged product.
This section illustrates the application of SEEM in an industrial  Spare parts provision. ABB provides to its distributors and end-
context. The aim is to provide insights on how the methodology customers a set of spare parts that could be ordered and shipped
has been used to re-engineer the ABB service portfolio. to the final destination.

The company The identification of the main customer needs is the second step
of the SEEM application to the service portfolio re-engineering.
ABB is a global leader in power and automation technologies. The Customer needs have been drawn out from marketing and
company is divided in five divisions that are, in turn, organized in customer’s data already available in ABB, and segmented by using
specific business units in relation to the customers and industries cluster analysis. The obtained segments have been described by
they serve. The ABB product portfolio is composed of complex using the persona model [44]. Among the key learning from this
offerings such as medium and high voltage power products, power marketing-oriented analysis, there is the evidence that ABB has to
systems, solutions for industrial processes optimization, discrete deal with heterogeneous types of customers with different needs
automation products, and low voltage products for electrical and expectations from ABB service. For the current analysis, two
application. As it could be seen, this diversified product portfolio categories with distinctive features have been taken into account:
needs different service requirements, i.e. features, price and lifecycle
intervention. The ABB service organization provides to its customers  Customers type I: these customers do not have internal
11 service categories ranging from traditional corrective mainte- capabilities to manage their maintenance activities and they
nance to system performance management. Such an extensive completely rely on ABB to maintain their installed base in a good
product-related service offering and heterogeneous product portfo- operating condition;
lio make the sharing of best practices among the business units a  Customers type II: these customers are usually large companies
daunting task. The adoption of a systematic service engineering with an internal team dedicated to maintenance. They directly
methodology along all the business units is crucial to properly take care of their installed base, and resort to ABB support only
identify and engineer PSS leading to increase service revenues and for complex service jobs and for critical spare parts.
fulfil customer needs. This motivated the implementation of the
SEEM at ABB. The research conducted highlighted that both these customers’
types share the same need, which is to maximize the availability of
SEEM application at ABB their installed base (‘‘maximize availability’’ in short hereafter).
This need has been the starting point for building the Service
The SEEM application in ABB deals with the re-engineering of Requirement Tree (SRT).
the actual product-related service portfolio of one specific business
unit. The rationale behind this decision is threefold. Firstly, this Company area: process prototyping
business unit is composed of more than fifty service-related Requirements design is the first task of the ‘‘process prototyp-
employees, making difficult for the service manager to assess ing’’ phase. It entails the development of the SRT to identify the
quickly how balanced is its business. Secondly, the product-related design requirements, and the implementation of the QFD logic to
service portfolio is complex and with some efficiency issues. weigh the activities and resources relevance in fulfilling customer
Finally, the customer segments are highly diversified. needs. Once the PSS(s) has been identified, the second task is the
The following paragraphs describe in details all the steps and designing the associated service delivery process(es).
the results obtained during the industrial application. For each
step, several meetings with ABB managers and employees,
involved in the service design and delivery, have been held to Requirements design: definition of the SRT. The Service Requirement
understand the customer needs, analyze the current offering and Tree (SRT) has been built for both the two types of customer
processes and to collect all the data. This allowed the research team starting from the common need ‘‘maximize availability’’. It is
26 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

important to highlight the effort spent by ABB managers to define managers and sales and marketing people. The weights have been
both the existing and hypothetical design requirements (DRs), assigned considering both the personal expertise and internal
activities (As) and resources (Rs). The single initial need allowed customer analysis data (e.g. ‘‘Net Promoter Score’’). Then, for each
the identification of three main wishes (optimize plant, reduce level of the SRT, the following QFD matrixes have been developed:
downtime and extend equipment lifetime), and 13 DRs. 20 differ- (i) need-wishes, (ii) wishes-DRs, (iii) DRs-activities, (iv) activities-
ent activities have been identified (e.g. manage order, handle resources. Table 5 shows an excerpt of the wishes-DRs matrix for
customer request). All the resources involved in delivering the one kind of customer.
services to the customers have been listed including humans (such Following the procedure reported in Service Engineering in the
as order handler or sales people), IT resources (such as ERP product service system context section, the resources’ relevance in
systems), and product components. maximizing equipment availability for both types of customers
An extract of the SRT is depicted in Fig. 3. resulting from the QFD based analysis is summarized in Fig. 4. This
The above figure shows the deployment of the need through the chart shows the resources to which ABB should pay particular
wish ‘‘reduce downtime’’. The company can support the customer attention while engineering and re-engineering its service delivery
in achieving this goal through, for example, the provision of proper processes in order to satisfy the needs of the two different kind of
installation, remote monitoring-based maintenance and commis- customers. In fact, it is fundamental to select the resources with
sioning (DRs). To offer these services, the activities that should be the right skills and avoid over utilization of such resources since
performed are ‘‘manage order’’, ‘‘handle customer request’’ and they have to manage the relation with the customer.
‘‘perform service job’’ (As). For a good order processing activity, the According to the QFD based analysis, the most important
‘‘sales people’’ and the ‘‘order handler’’ are the relevant resources resources to satisfy customer type II are ‘‘sales people’’, ‘‘training
that should be employed. The DR ‘‘remote monitoring-based operators’’, ‘‘spare parts’’ and ‘‘warehouse operators’’. This result is in
maintenance’’ is connected to a ‘‘monitoring’’ activity and to a agreement with the definition of the customer type II: if he/she
product component (i.e. the sensors). As previously stated, the wants to perform the maintenance activities on his/her own, he/she
design of product components is left to traditional methods and would definitely need a good training and a fast spare parts delivery
tools which are already implemented in the company. service for which the ‘‘warehouse operator’’ and the ‘‘sales people’’
are key players. On the other hand, to satisfy customer type I, who
Requirements design: QFD based analysis. After the definition of the completely relies on ABB, the most critical resources are the ‘‘sales
SRT, a QFD based analysis has been carried out. To this purpose, a people’’ who defines the contract terms and conditions, and the
weight (1, 3 or 9) has been assigned to each branch of the SRT in ‘‘technicians’’ who perform the service job. These results about
order to define the most critical activities and the associated resources’ relevance have been the key levers to identify possible
resources in complying with customer’s need. Theoretically, this process improvement, keeping a high focus on customer needs.
activity should be ‘‘co-developed’’ with the customer. However, If needed, the QFD based analysis can be adopted to support the
obtaining customers availability is complex and takes a lot of time classification of DRs and activities according to their relevance for
and effort. Therefore, the evaluation has been carried out by service customers. Thus, it would help in defining company’s strategy.

Fig. 3. Excerpt from the ABB SRT.

Table 5
Excerpt of the ABB wishes-DRs matrix.

Design requirements Wishes DR importance DR importance %

Reduce downtime Extended equipment lifetime

Corrective maintenance 3 – 27 4
Commissioning and installation 3 – 27 4
Remote monitoring-based maintenance 9 3 108 17
Wishes weights 9 3
 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32 27

Fig. 4. Resources relevance for the two customer types in maximizing equipment availability.

Process design. The second task of the process prototyping is process this task, also selects the technician(s) to perform the service job
design using the blueprinting technique [57,59]. Considering the focus according to staff availability and to failure criticality.
of the case, four different service delivery processes have been  Prepare service job. In this phase, the technicians define the spare
analyzed and drawn in a blueprint map using MS Visio. In these maps, parts and the material needed for the intervention. In the case of
all the activities have been represented, putting into evidence the workshop maintenance, the customer sends the product to ABB’s
resources performing them and their relation with the customer premise.
process. About 120 activities have been identified for each service  Perform service job. In the case of onsite maintenance, the
delivery process performed by the customers, ABB front-end resources ‘‘technician(s)’’ go(es) to the customer, whereas in the other case
(e.g. sales people, onsite technicians), ABB backstage resources (order the ‘‘production workers’’ perform the service job, or assemble
handlers, workshop technicians), and support processes employees the retrofitting kit or the spare parts at the ABB facility.
(e.g. logistics, administration). The service delivery processes of the  Complete service job. The final part of the process entails the
ABB service offering are characterized by a common structure that is shipment of the materials to the customer (in case of workshop
described hereafter and represented in Fig. 5: maintenance) and the collection of all the documents. Finally, the
invoice is sent to the customer.
 Handle customer request. The process starts with a service request
from the customer. The sales people receive these requests, The next section focuses on the validation of the service
analyze the customer reliability and define a quotation for the delivery process devised on the blueprint.
selected service.
 Confirm capability. The customer reviews the ABB offer and Company area: process validation
determines whether it fits its requirements. Then, the customer In a re-engineering case, the main goal of the process validation
sends a service order to ABB. step is the assessment of the current processes’ performance and
 Manage order. Once the service order is received, it is compared the identification of the resource configuration that better balance
to the offer to check its alignment, and then uploaded in the the external performance (i.e. customer satisfaction based on cycle
ABB’s ERP system with all the related information. time) and internal efficiency.
 Mobilize and plan. This phase strictly refers to the case of Considering the mid and long term ABB strategies and targets
intervention at the customer’s plant. ABB and the customer agree for the service business, the as-is future target scenario has been
on a date to perform the service and set all the documentation defined and the what-if analysis has been carried out in order to
needed before the intervention. The ‘‘dispatcher’’, responsible for evaluate the performance of current organization. Based on these

Fig. 5. ABB Service Blueprinting map.

28 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

results, the best solution has been identified among the different 2. Customer satisfaction based on cycle time. This performance has
scenarios ensuring a balance between the external performance, been measured through the indicator proposed by the SEEM and
evaluated through the customer satisfaction based on cycle time, presented in Process validation section. The data used as input
and the internal performance. For this purpose, the Discrete Event to calculate this indicator are standard output of the simulation
Simulation (DES) approach has been adopted. The blueprinting software.
maps represented in Microsoft Visio have then been translated into
a simulation model with ProModel Process Simulation software. Following the validation procedure for the re-engineering case,
The simulation model has been developed considering the the model, simulating the current process, has been run (as-is
entities as the customer requests of the different services and the scenario) and refined until the company has confirmed its
events as the service delivery process activities. The entry alignment with the reality.
distributions of the entities have been inferred from ABB historical The ABB forecasts (not reported for privacy reasons) have been
data by calculating the best fitting distribution. set into the model to define the as-is future target scenario. The as-is
With regard to activities’ duration, a distribution function has future target analysis showed how ABB would face the forecasted
been specified, according to the available data or to ABB employ- changes with the current organization. In particular, in the as-is
ee’s experience. The majority of them have been set as a triangular future target data related to the introduction of new product-
distribution with a minimum, a maximum and a mode. Further- related services and the increase and decrease of the demand for
more, a resource or a group of resources has been assigned to each some services have been evaluated. As expected, those changes
activity. Additionally, for each human resource, the working affect all the process parameters and in particular the service cycle
schedule has been appointed to define the amount of their time time and the resource utilization.
dedicated to the service process under investigation. In total, The as-is future target simulation results showed three main
55 different resources have been modelled. bottlenecks (in terms of long queues and high resource utilization):
Once validated, the simulation has been run over a period of (i) handle customer request (performed by the ‘‘sales people’’), (ii)
three years for ten replications to ensure consistency and mobilize and plan, referred to the service jobs performed at the
variability of the results. Moreover, since the system at time zero customer site (‘‘dispatcher’’), and (iii) performing service job
has been assumed empty, a warm-up time of six months has been (‘‘technicians’’).
set. The results obtained in terms of bottlenecks, resource utiliza-
At the end of the simulation, the results have been collected and tion, customer satisfaction based on cycle time and some input
compared with real data to check the robustness of the model. The from the ABB future strategy have been considered as the starting
main KPIs, belonging to the two above-mentioned categories, have point to develop the scenarios of the what-if analysis.
been the following: In particular, the scenarios of the what-if analysis have been
identified combining the factors predominantly influencing the
1. Company internal measures: process performance. One of the main influencing factors is
– Number of completed service jobs per year. For each kind of represented by the capacity of the ‘‘sales people’’, the ‘‘dispatcher’’
service, the number of requests yearly received, together with and the ‘‘technicians’’, which have been revealed critical in the as-is
those completed have been identified. The number of entities future target and significant in the QFD based analysis for the
still in the system or exiting from the system is automatically customer. In addition, with regard to the bottlenecks identified in
shown by the simulation software; the ‘‘handling customer request’’ and ‘‘manage order’’ activities,
– Time to complete a service job. The time for processing each the automation of such activities through the introduction of IT
kind of service request has been measured with a ‘‘ad hoc’’ systems has been identified as a further factor influencing the
function monitoring the time laps between the request arrival overall process performance.
and its conclusion; Thus, for each influencing factor, possible alternatives
– Resource utilization. It represents the utilization of the human (Value) have been identified with the ABB service managers
resources based on the time they dedicate to the process and are reported in Table 6 together with the notation that will
activities. Resource utilization is a standard output of the be used (Set of value notation). For example, in relation to the
simulator; ‘‘handle customer request’’ factor, two alternatives have been
– Queues: it has been measured as the waiting time to perform identified:
an activity and it is a standard output of the simulation
software. The queue lengths along with the resource  introduction of new IT tools and related procedures to automate
utilization have been used to identify the bottlenecks in the some process activities;
system.  keeping the process unchanged.

Table 6
Factors and response value for the development of the scenarios.

Factors Type of value Value Set of value notation

Handle customer request (hcr) Qualitative hcr = s: Automated process (referred to as hcrs in the following) HCR
hcr = n: Non automated process (hcrn)
Manage order (mo) Qualitative mo = s: Automated process (mos in the following) MO
mo = n: Non automated process (mon)
Sales people capacity change (pc) Quantitative pc = 0: current capacity of the sales people (pc0) PC
pc = 12: +12 h sales people (pc12)
pc = 16: +16 h sales people (pc16)
Dispatcher capacity change (dc) Quantitative dc = 2: +2 h Dispatcher (dc2) DC
dc = 4: +4 h Dispatcher (dc4)
dc = 6: + 6 h Dispatcher (dc6)
Technicians capacity change (tc) Quantitative tc = 24: +24 h Technicians (tc24) TC
tc = 32: +32 h Technicians (tc32)
 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32 29

Each scenario is a combination of the factor values, and can be proposed in the scenarios A and B, the utilization does not exceed
written as: the 80% threshold. This threshold is feasible considering possible
extra working hours that have not been set in the resource
SC t ¼ ðhcr; mo; pc; dc; tcÞ 8t2T (7) scheduling in the simulation model.
Furthermore, the what-if analysis is relevant to see how changes
where T is the set of all the scenarios. Considering the five factors
in the service portfolio affect the company organization. In fact, the
and their possible value, 72 scenarios to be experimented can be
‘‘production workers’’ utilization from the as-is to the as-is future
identified, that is:
target scenario drastically decreases due to the removal of one
jTj ¼ jHCRjjMOjjPCjjDCjjTCj ¼ 72T (8) service from the offer. Based on this result, the company should
define substitutive activities for this kind of human resource.
where the operator |x| returns the cardinality of the set X. Regarding service cycle time, the time needed to handle customer
All the 72 identified scenarios have been qualitatively analyzed requests and to perform the service job has been thoroughly analyzed
by the researchers and the ABB managers. Among them, since they were the bottlenecks of the process. Fig. 7 reports the
16 scenarios have been considered feasible and representative results of the what-if analysis. It emerged that, in the as-is future
of the reality. In this paper, only the two most significant scenarios target, the total service cycle time increases significantly. However, in
are reported for brevity, referred to as scenario A and B. Table 7 scenario A and B this time is lowered to the actual level (for privacy
reports the values assumed by each factor in the two selected reason, the values on the Y-axis cannot be reported).
scenarios along with the data describing the changes made to the In addition, the results related to customer satisfaction based on
processes. For example, in Scenario A, the automation of the cycle time have been analyzed. The indicators SAhk have been
process activities causes a 50% reduction of the time to define a calculated for each activity and design requirement as described in
standard offer and a 25% reduction of the development time of a Process validation section. TtargetA(h) has been fixed the minimum
complex offer. activity duration obtained during the as-is simulation, since the
These two scenarios have been selected since they allow company does not have fixed ad-hoc target duration for each
performance improvement both from internal and external point activity. Concerning the customer satisfaction based on cycle time
of view. For each kind of product-related service under analysis, (Table 8), the overall indicator (S) related to customer type I
the improvement actions suggested in these scenarios helped to segment is 85.46% in the as-is model, then it decreases to 79.48% in
achieve acceptable total duration (aligned with the actual one), the as-is future target model and finally it increases to 91% in
proper resource utilization (lower than 80%) and an adequate scenarios A and B. The same trend is observed in the customer type
customer satisfaction based on cycle time (comparable or higher II segment case, where S is equal to 95.90% in the as-is model, to
than the current one). 94.80% in the as-is future target model and to 96.99% (Scenario A)
In Fig. 6, the average utilization of the most relevant groups of and 97.16% (Scenario B). These trends are due to changes of the
resources is depicted. As it is possible to observe, in the as-is future activity duration since the activities importance rates have been
target, the sales people and the dispatcher reach a 100% utilization kept unchanged. Therefore, as expected, it is possible to argue that
that is not feasible, in reality, for human being. With the changes reducing the total service cycle time, the overall customer

Table 7
Description of analyzed scenarios.

Scenario A Scenario B
SCA = (hcrs, mos, pc0, dc2, tc24) SCB = (hcrn, mon, pc12, dc6, tc32)

Change in the process

Change in the proposal process Automated process (hcrs) involving: Non automated process (hcrn)
- 50% reduction of the time to define a standard offer
- 25% reduction of the development time of a complex offer
Change in the analysis Automated process (mos) involving: Non automated process (mon)
of the order - Reduction of 50% of the time to check order coherence
with the proposal
Change in the capacity of resource (working hours)
Sales people capacity pc0 – No change pc12 – Increase of 12 working hours per day
Dispatcher capacity dc2 – Increase of 2 working hours per day dc6 – Increase of 6 working hours per day
Technicians capacity tc24 – Increase of 24 working hours per day tc32 – Increase of 32 working hours per day

Fig. 6. Scheduled utilization of Resources.

30 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

Fig. 7. Average total cycle time for design requirements.

Table 8
Customer satisfaction based on cycle time indicators.

Activities Customer satisfaction based on cycle time indicators S

As-is As-is future target Scenario A Scenario B

Customer Customer Customer Customer Customer Customer Customer Customer

type I (%) type II (%) type I (%) type II (%) type I (%) type II (%) type I (%) type II (%)

S 85.46 95.90 79.48 94.80 91.14 96.99 91.96 97.16

satisfaction based on cycle time increases. This phenomenon is re-engineer the service delivery process providing a useful support
observed in the case of the improved scenarios A and B, where the while making decision. More in detail, according to the case
total service cycle time is significantly reduced and, consequently, presented and to other applications in different ABB business units,
the increase of the overall customer satisfaction based on cycle the SEEM has been used to set PSSs in terms of service delivery
time, as considered in the SEEM, is respectively 6% for customer’s process capable of balancing internal and external performance
type I and 1% for customer’s type II. through:
The customer satisfaction based on cycle time indicator could
help the company in defining and selecting the best possible (a) a systematic evaluation of internal and external performance of
process configuration. A cross analysis considering the number of the as-is process;
resources involved, their utilization and S could be a good way to (b) the analysis of possible balance between internal and external
balance the external with the internal performance. For example, performance;
the selection between scenario A and scenario B can be the (c) the comparison of a variety of service delivery configurations.
strategic choice between a slightly higher customer satisfaction
based on cycle time achievable in B and better resource utilization In terms of management implication, the test case demon-
in scenario A. The company could decide whether to focus on the strates the validity of the SEEM in supporting all that phases of PSS
customer satisfaction based on cycle time while another one can engineering with a deeper focus on the service delivery process.
believe that for such a small increase in the customer satisfaction According to managers’ feedback, the adoption of such approach
based on cycle time, a lower resource utilization, and the related created shared awareness about current processes and inefficien-
increase in terms of cost, is not justified. cies. In addition, it supported the definition of a new process
Summarizing, the what-if analysis provides some suggestions characterized by a better resources planning and a higher
that should be taken into account when the company is planning to efficiency in dealing with customer requests. Moreover, the joint
change its product-related service offering or the customer analysis of resources utilization and customer satisfaction based
requests are expected to change. The results emerged represent on cycle time make managers capable of taking structured and
just a possible way to solve future issues related to services, and justified decisions.
they would be used as hints to balance the service organization to For what regards the methodology itself, the process validation
the future needs. phase revealed as the most time consuming phase due to the
following differences between the static and the dynamic maps:
Discussion
 Unique simulation model. Since the different service processes
The test case presented in the previous section demonstrated analyzed share several resources, such as sales people, techni-
the robustness of the SEEM in being applied to an industrial case. In cians and order handlers, a single simulation model is required.
particular, the step by step methodology can support company The amount of time that each resource dedicates to a specific
managers to: (i) guide the identification of customer needs, (ii) service could not be defined a priori, since it depends on many
identify possible PSSs fulfilling such needs, (iii) identify the factors such as the period of the year, the priority of each request
company activities and resources to implement such PSSs and (iv) and the specific intervention.
 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32 31

 Level of detail. The service blueprinting maps present thepro- more generalized and mature theoretical framework would finally
cesses in a very detailed way. In the simulation model, such a help in developing proper integrated tools.
detailed representation may be a problem, since the duration and
time variability must be included when setting the process Acknowledgements
parameters. Setting the time for many detailed activities increases
significantly the variability at levels that do not reflect reality. In This research has been partly funded by ABB research grant and
order to avoid this problem, it is crucial to group together some the European Commission through the ProSSaLiC (GA-2010-
activities that are sequential and logically linked and that, 269322). The authors gratefully acknowledge extremely valuable
together, can become a macro-process in the final model. collaboration of all ABB managers and scientists involved in the
 Hierarchical structure. The simulation model, due to the huge project.
number of activities involved, required a lot of time to be set. To
facilitate the sharing and the comprehensiveness of the model, it
References
has been represented with a two level hierarchical structure. The
two levels have been designed according to Rondini et al. [79] to [1] Gaiardelli, P., Martinez, V., Cavalieri, S., 2015, The Strategic Transition to
obtain a suitable overview of the company and customer Services: A Dominant Logic Perspective and Its Implications for Operations,
performance. The first level provides the sequence of the main Production Planning & Control, 26/14–15: 1165–1170.
[2] Aurich, J., Mannweiler, C., Schweitzer, E., 2010, How to Design and Offer
activities (A), identified in the SRT, while in the second level a Services Successfully, CIRP Jorunal of Manufacturing Science and Technology,
breakdown in terms of sub-activities is detailed. 2:136–143.
[3] Cavalieri, S., Pezzotta, G., 2012, Product-Service Systems Engineering: State of
the Art and Research Challenges, Computers in Industry, 63/4: 278–288.
[4] Mont, O., 2002, Clarifying the Concept of Product-Service System, Journal of
Conclusions Cleaner Production, 10/3: 237–245.
[5] Berry, L., Carbone, L., Haeckel, S., 2002, Managing the Total Customer Experi-
ence, MIT Sloan Management Review, 85–90.
Industrial companies are facing the need of tools and methods [6] Cavalieri, S., Pezzotta, G., Shimomura, Y., 2012, Product-Service System Engi-
to design and assess their PSS offering and the related service neering: from Theory to Industrial Applications, Computers in Industry, 63:
delivery processes. This study presents the SEEM methodology 275–277.
[7] Van Halen, C., Vezzoli, C., Wimmer, R., 2005, Methodology for Product Service
that assists companies in balancing the company internal and System Innovation: How to Develop Clean, Clever and Competitive Strategies
external performance while (re)-engineering its PSS offering. The in Companies, Uitgeverij Van Gorcum, Assen, Netherlands.
methodology, which is composed of two main areas (company and [8] Alonso-Rasgado, T., Thompson, G., Elfström, B., 2004, The Design of Functional
(Total Care) Products, Journal of Engineering Design, 15/6: 515–540.
customer), has been applied also to an industrial case to illustrate
[9] Baines, T., Lightfoot, H., Evans, S., Neely, A., Greenough, R., Peppard, J., Roy, R.,
its applicability. The use case, focused on the re-engineering of Shehab, E., Braganza, A., Tiwari, A., 2007, State-of-the-Art in Product-Service
service portfolio at ABB business unit, demonstrates the complex Systems, Proceedings of the Institution of Mechanical Engineers, 221/10:
definition of a PSS solution and the critical capability planning (i.e. 1543–1552.
[10] Kett, H., Voigt, K., Scheithauer, G., Cardoso, J., 2008, Service Engineering in
resources, tools and spare parts) in service delivery. Business Ecosystems, Proc. XVIII International RESER Conference (Stuttgart,
The industrial case showed its appropriateness and robustness Germany).
to identify possible PSS solutions and to address the complexity of [11] Rapaccini, M., Saccani, N., Pezzotta, G., Burger, T., Ganz, W., 2013, Service
Development in Product-Service Systems: A Maturity Model, Service Indus-
assessing the performance of the service delivery of PSS offerings. tries Journal, 33/3-4: 300–319.
In particular, the main benefits of this methodology, demon- [12] Aurich, J., Fuchs, C., Wagenknecht, C., 2006, Modular Design of Technical
strated with the application on ABB, are: (i) the adoption of a Product–Service Systems, Innovation in Life Cycle Engineering and Sustain-
able Development, Springer Netherlands, The Netherlands: 303–320.
systematic procedure to analyze the existing portfolio; (ii) the [13] Shimomura, Y., Tomiyama, T., 2005, Service Modeling for Service Engineering,
improvement of the delivery performance by the identification of IFIP International Federation for Information Processing, 167:31–38.
resources or service activities directly affecting customer needs, [14] Maussang, N., Zwolinski, P., Brissaud, D., 2009, Product–Service System Design
Methodology: From the PSS, Journal of Engineering Design, 20/4: 349–366.
(iii) a better definition of the process changes in order to properly [15] Moser, U., Maisenbacher, S., Kaspereka, D., Maik, M., 2015, Definition of an
manage an increase/decrease of demand or changes in service Approach for the Development of Product-Service Systems, Procedia CIRP,
portfolio, and (iv) the definition of a possible service delivery 30:18–23.
[16] Pezzotta, G., Cavalieri, S., Gaiardelli, P., 2012, A Spiral Process Model to
process able to create at the same time value for the customer and
Engineer a Product Service System: An Explorative Analysis through Case
profitability for the company. Studies, CIRP Journal of Manufacturing Science and Technoloy, 5/3: 214–225.
In parallel to the benefits, the application to a case shed the light [17] Geng, X., Chu, X., Xue, D., Zhang, Z., 2010, An Integrated Approach for Rating
on possible improvement of it. Engineering Characteristics’ Final Importance in Product-Service System De-
velopment, Computer in Industrial Engineering, 59/4: 585–594.
The main limitation of the methodology is related to the [18] Barquet, A., Gouvea, M., Roman, C., Pinheiro, V., Rozenfeld, H., 2013, Employing
customer satisfaction based on cycle time indicator. The main the Business Model Concept to Support the Adoption of Product-Service
assumption behind this application is that the minimum cycle time Systems (PSS), Industrial Marketing Management, 42/5: 693–704.
[19] Froehle, C., Roth, A., Chase, R., Voss, C., 2000, Antecedents of New Service
of the current service delivery process is the optimal one. However, Strategic Operations Choices, Journal of Service Research, 3/3: 3–17.
to improve the methodology it could be useful to identify an [20] Luczak, H., Gill, C., Sander, B., 2007, Architecture for Service Engineering – The
optimal time interval for both customer and company sides in Design and Development of Industrial Service Work, Advances in Services
Innovations, Springer Berlin Heidelberg: 47–63.
order to achieve a more meaningful measure. The definition of this [21] Bullinger, H.-J., Fähnrich, K.-P., Meiren, T., 2003, Service Engineering – Me-
interval could help organizations to distinguish those activities thodical Development of New Service Products, International Journal produc-
that have low levels of customer satisfaction based on cycle time tion Economics, 275–287.
[22] Kimita, K., Shimomura, Y., Arai, T., 2009, Evaluation of Customer Satisfaction
indicator and that have to be improved.
for PSS Design, Journal of Manufacturing Technology Management, 20/5:
In addition, so far the second step of the methodology focuses 654–673.
only on the service part of the PSS and the design of product [23] Arai, T., Shimomura, Y., 2004, Proposal of Service CAD System – A Tool for
Service Engineering, CIRP Annals, 53/1: 397–400.
components are left to traditional product design methods. The
[24] Sundin, E., Lindahl, M., Sakao, T., Shimomura, Y., Björkman, M., 2005, New
methodology could be improved with a better integration of Engineering Design for Functional Sales Business, Melbourne.
product design also during the second step of the framework. [25] Lindahl, M., Sundin, E., Shimomura, Y., Sakao, T., 2006, An Interactive Model for
Future works will be related to the adoption of the methodology Service Engineering of Functional Sales Offers, Leuven.
[26] Sakao, T., Shimomura, Y., 2007, Service Engineering: A Novel Engineering
in other cases and in other industries in order to test extensively its Discipline for Producers to Increase Value Combining Service and Product,
applicability and to further generalize the SEEM. Indeed, having a Journal of Cleaner Production, 15/6: 590–604.
32 G. Pezzotta et al. / CIRP Journal of Manufacturing Science and Technology 15 (2016) 19–32

[27] Tukker, A., 2015, Product Services for a Resource Efficient and Circular Econo- [54] Snodgrass, T., Fowler, T., 1972, Customer Oriented FAST Diagramming, Pro-
my – A Review, Journal of Cleaner Production, 97:76–91. ceeding of Society of American Value Engineers Conference.
[28] Beuren, F.H., Ferreira, M., Miguel, P., 2013, Product-Service Systems: A Liter- [55] Maussang, N., Sakao, T., Zwolinski, P., Brissaud, D., 2007, A Model for Designing
ature Review on Integrated Products and Services, Journal of Cleaner Produc- Product-Service Systems Using Functional Analysis and Agent Based Model,
tion, 47:222–231. International Conference on Engineering Design, 20/4: 349–366.
[29] Boehm, M., Thomas, O., 2013, Looking Beyond the Rim of One’s Teacup: A [56] Geum, Y., Park, Y., 2011, Designing the Suistainable Product-Service Integra-
Multidisciplinary Literature Review of Product-Service Systems in Information tion: A Product-Service Blueprint Approach, Journal of Cleaner Production, 19/
Systems, Business Management and Engineering Design, Journal of Cleaner 14: 1601–1614.
Production, 51:245–260. [57] Shostack, G., 1982, How to Design a Service, European Journal of Marketing,
[30] Qu, M., Yu, S., Chen, D., Chu, J., Tian, B., 2016, State of the Art of Design, 16/1: 49–63.
Evaluation, and Operation Methodologies in Product Service Systems, Com- [58] Shostack, G., 1984, Designing Services That Deliver, Harvard Business Review,
puters in Industry, 77:1–14. 62/1: 133–139.
[31] Hara, T., Arai, T., Shimomura, Y., Sakao, T., 2009, Service CAD System to [59] Bitner, M.J., Ostrom, A., Morgani, F., 2008, Service Blueprinting: A Practical
Integrate Product and Human Activity for Total Value, CIRP Journal of Technique for Service Innovation, California Management Review, 50/3: 66–94.
Manufacturing Science and Technology, 1/4: 262–271. [60] Bianchi, N., Evans, S., Revetria, R., Tonelli, F., 2009, Influencing Factors of
[32] Morelli, N., 2002, Designing Product/Service Systems: A Methodological Ex- Successful Transitions towards Product-Service Systems: A Simulation Ap-
ploration, MIT Design Issues, 18/3: 3–18. proach, International Journal of Mathematics and Computers in Simulation, 1/
[33] Alix, T., 2010, A Framework for Product-Service Design for Manufacturing 3: 30–43.
Firms, IFIP WG 5.7 International Conference on Advances in Production [61] Pezzotta, G., Rondini, A., Pirola, F., Pinto, R., 2016, Evaluation of Discrete Event
Management Systems: New Challenges, New Approaches. Simulation Software to Design and Assess Service Delivery Processes, Service
[34] Marques, P., Cunha, P.F., Valente, F., Leitao, A., 2013, A Methodology for Supply Chain Systems: A Systems Engineering Approach, (in press).
Product-Service System Development, Procedia CIRP, 7:371–376. [62] Maisenbacher, S., Weidmann, D., Kasperek, D., Omer, M., 2014, Applicability of
[35] Sandin, J., Berggren, C., 2015, A Simplified Service Engineering Approach Used Agent-Based Modeling for Supporting Product-Service System Development,
by an Industrial Service, Procedia CIRP, 30:30–35. Procedia CIRP, 16:356–361.
[36] Trevisan, L., Lelah, A., Brissaud, D., 2015, New PSS Design Method of a [63] Hirth, N., Maisenbacher, S., Kasperek, D., Hollauer, C., Maurer, M., 2015, An
Pneumatic Energy System, Procedia CIRP, 30:48–53. Approach to Reveal Starting Point for PSS Design Support with Dynamic
[37] Marilungo, E., Peruzzini, M., Germani, M., 2015, An Integrated Method to Models, Procedia CIRP, 30:462–467.
Support PSS Design within the Virtual Enterprise, Procedia CIRP, 30:54–59. [64] Shimomura, Y., Hara, T., Arai, T., 2008, A Service Evaluation Method Using
[38] Visintin, F., Porcelli, I., Ghini, A., 2014, Applying Discrete Event Simulation to Mathematical Methodologies, CIRP Annals – Manufacturing Technology, 57/1:
the Design of a Service Delivery System in the Aerospace Industry: A Case 437–440.
Study, Journal of Intelligent Manufacturing, 25/5: 1135–1152. [65] Bertoni, A., Bertoni, M., Isaksson, O., 2013, Value Visualization in Product
[39] Chalal, M., Boucher, X., Marques, G., 2015, Decision Support System for Service Systems Preliminary Design, Journal of Cleaner Production, 53:
Servitization of Industrial SMEs: A Modelling and Simulation Approach, 103–117.
Journal of Decision Systems. [66] Luczak, H., Gill, C., Sander, B., 2007, Architecture for Service Engineering – The
[40] Medini, K., Boucher, X., 2016, Value Chain Configuration for PSS delivery – Design and Development of Industrial Service Work, Advances in Services
Evidence from an Innovative Sector for Sludge Treatment, CIRP JMST, 14–24. Innovations, Berlin/Heidelberg: 47–63.
[41] Yoon, B., Kim, S., Rhee, J., 2012, An Evaluation Method for Designing a New [67] Chuang, P.-T., 2007, Combining Service Blueprint and FMEA for Service Design,
Product-Service System, Expert System with Application, 39:3100–3108. The Service Industries Journal, 27/2: 91–104.
[42] Pezzotta, G., Pirola, F., Pinto, R., Akasaka, F., Shimomura, Y., 2015, A Service [68] Lee, H., Kim, C., Park, Y., 2010, Evaluation and Management of New Service
Engineering Framework to Design and Assess an Integrated Product-Service, Concepts: An ANP-Based Portfolio Approach, Computers & Industrial Engi-
Mechatronics . neering, 58/4: 535–543.
[43] Vincent, T., Selvarani, D., 2013, Personal Values Approach For a Better Under- [69] Lee, S., Geum, Y., Lee, S., Park, Y., 2015, Evaluating New Concepts of PSS Based
standing of Consumer Behaviour, International Journal of Innovative Research on the Customer Value: Application of ANP and Niche Theory, Expert Systems
& Development, 2/2: 509–517. with Applications, 42:4556–4566.
[44] Pirola, F., Pezzotta, G., Andreini, D., Galmozzi, C., Savoia, A., Pinto, R., 2014, [70] Hu, H., Chen, S., Hsu, C., Wang, C., Wu, C., 2012, Development of Sustainability
Understanding Customer Needs to Engineer Product-Service Systems, Evaluation Model for Implementing Product Service Systems, International
Advances in Production Management Systems. Innovative and Knowledge- Journal of Environmental Science and Technology, 9/2: 343–354.
Based Production Management in a Global-Local World, Springer Berlin [71] Sakao, T., Shimomura, Y., Sundin, E., Comstock, M., 2009, Modeling Design
Heidelberg, Ajaccio, France. Objects in CAD System for Service/Product Engineering, Computer-Aided
[45] Shimomura, Y., Akasaka, F., 2013, Toward Product-Service System Engineer- Design, 41:197–213.
ing: New System Engineering for PSS Utilization, in Meier H A, (Ed.) Product- [72] Ganjar, A.J.R., Byungun, Y., 2014, A Simulation Tool for Prioritizing Product-
Service Integration for Sustainable Solutions. Springer, pp. pp.27–40. Service System (PSS) Models in a Carsharing Service, Computers & Industrial
[46] Hara, T., Arai, T., Shimomura, Y., 2009, A Method to Analyze PSS from the Engineering, 70:59–73.
Viewpoint of Function Service Activity and Product Behavior, 1st CIRP IPS2 [73] Abu-Taieh, E., El Sheikh, A., 2007, Commercial Simulation Packages: A Compara-
Conference, 180. tive Study, International Journal of Simulation, 8/2: 66–76.
[47] Aurich, J.C., Fuchs, C., Wagenknecht, C., 2006, Life Cycle Oriented Design [74] Harpring, R., Evans, G.W., Barber, R., Deck, S.M., 2014, Improving Efficiency in
of Technical Product–Service Systems, Journal of Cleaner Production, Social Services with Discrete Events Simulation, Computers and Industrial
14:1480–1494. Engineering, 70:159–167.
[48] Wang, P., Ming, X., Li, D., Kong, F., Wang, L., Wu, Z., 2011, Modular Develop- [75] Laughery, K.R., Laughery, B.R., Lovvoll, D.R., Mcquilkin, M.L., Wogalter, M.S.,
ment of Product Service Systems, Concurrent Engineering: Research and 1998, Effects of Warnings on Responsibility Allocation, Psychology & Market-
Applications, 19:85–96. ing, 15:687–706.
[49] Kim, S., Yoon, B., 2012, Developing a Process of Concept Generation for New [76] Rondini, A., Pirola, F., Fanelli, A., Spini, A., Pezzotta, G., Ouertani, M.-Z., 2015,
Product-Service Systems: A QFD and TRIZ-Based Approach, Service Business, The Evaluation of Customer Satisfaction for Product-Service Systems: An
6/3: 323–348. Indicator to Integrate the Service Engineering Methodology, Proceeding of
[50] Shimomura, Y., Hara, T., Arai, T., 2009, A Unified Representation Scheme for Summer School ‘‘Francesco Turco’’, Naples (Italy).
Effective PSS Development, CIRP Annals – Manufacturing Technology, 58/1: [77] Clark, G.L., Kaminski, P.F., Rink, D.R., 1992, Consumer Complaints: Advice on How
379–382. Companies Should Respond Based on an Empirical Study, Journal of Consumer
[51] Nemoto, Y., Ueia, K., Sato, K., Shimomura, Y., 2015, A Context-Based Require- Marketing, 9/3: 5–14.
ments Analysis Method for PSS Design, Procedia CIRP, 30:42–47. [78] Mattila, A.S., Mount, D.J., 2003, The Impact of Selected Customer Character-
[52] Song, W., Ming, X., Han, Y., Wu, Z., 2013, A Rough Set Approach for Evaluating istics and Response Time on e-Complaint Satisfaction and Return Intent,
Vague Customer Requirement of Industrial Product-Service System, Interna- International Journal of Hospitality Management, 22/2: 135–145.
tional Journal of Production Research, 22/51: 6681–6701. [79] Rondini, A., Pezzotta, G., Pirola, F., Pinto, R., Ouertani, Z.M., 2014, Service
[53] Kim, S., Yoon, B., 2014, A Systematic Approach for New Service Concept Engineering Framework: The Adoption of Simulation to Design and Configure
Generation: Application of Agent-Based Simulation, Expert Systems with Product-Service Solutions, Proceeding of 21st ICE Conference, ‘‘Engineering
Applications, 41:2793–2806. Responsible Innovation in Products and Services’’, Bergamo.