UX- for Smart-PSS: Towards a Context-aware Framework

Angela Carrera-Rivera

1 a

, Felix Larrinaga

1 b

, Ganix Lasa

2 c

and Giovanna Martinez-Arellano

3 d

Faculty of Engineering, Mondragon Unibertsitatea, Loramendi 4, 20500 Arrasate, Spain

Design Innovation Center (DBZ), Mondragon Unibertsitatea, Loramendi 4, 20500 Arrasate, Spain

Faculty of Engineering, University of Nottingham, Nottingham NG7 2RD, U.K.

Keywords:

S-PSS, Context-awareness, User Experience, Recommendation System.

Abstract:

Smart-product service systems are a business strategy that combines product and service into one value propo-

sition. The user experience of digital services and the smart product can be a clear differentiator among

competitors to achieve economically sustainable solutions. Hence, offering a more personalized experience

is an important aspect of S-PSS. This paper aims to provide a theoretical framework for a context-aware user

experience in S-PSS by providing adaptive and personalized services to the users according to their needs

in a given context, by exploiting the digital capabilities of smart products and referring to the use of rec-

ommendation systems. The paper presents an application scenario using a smart-wearable as an example of

a product-oriented PSS to better describe the framework and each component while stating the future chal-

lenges.

1 INTRODUCTION

Smart-Product Service Systems (S-PSS) is an ecosys-

tem that proposes the combination of both product

and service into one value proposition. The design

of S-PSS has received increased attention in recent

years (Carrera-Rivera et al., 2022; Cong et al., 2020;

Dou and Qin, 2017) and implies several challenges.

From the service perspective, S-PSS offers digital ser-

vices that can be product independent or product de-

pendent, powered by the integrated sensors in smart

connected products or by technologies like digital

twin or augmented reality (AR) (Zheng et al., 2019b).

Zheng et al. (2019b) deﬁned S-PSS as “an IT-driven

value co-creation business strategy consisting of var-

ious stakeholders as the players, intelligent systems

as the infrastructure, smart, connected products as

the media and tools, and their generated e-services

as the key values delivered that continuously strives

to meet individual customer needs in a sustainable

manner”. However, the design of S-PSS in multi-

ple studies is only limited to the early phases of the

product development and they do not address the evo-

https://orcid.org/0000-0001-8593-5961

https://orcid.org/0000-0003-1971-0048

https://orcid.org/0000-0002-2424-5526

https://orcid.org/0000-0003-3105-4151

lution and adaptability that S-PSS should have. S-

PSS follows the Service-Dominant (S-D) logic, a fun-

damental pillar of Service Design, that characterizes

for a “value co-creation” view that uses experiences,

context and multi-stakeholder participation to create

innovative business value propositions(Wetter-Edman

et al., 2014). Users may have different roles as co-

creators in the different phases of S-PSS lifecycle:

co-ideators, co-innovators, co-evaluators, co-testers

or experience creators (Pezzotta et al., 2017). Users

as experience creators can be involved in the solu-

tion conceptualization supporting better customiza-

tion. Pezzotta et al. (2017) stated that providers can

generate richer experiences for customers by under-

standing their preferences. A deep understanding

of the customer experiences could help companies

in deﬁning and re-deﬁning better value propositions.

User Experience(UX) can be a clear differentiator

among competitors to achieve economically sustain-

able PSS. Cong et al. (2020) highlighted that “user

preferences should be associated with different design

elements of Smart PSS in speciﬁc usage contexts”.

Previous studies have focused their attention on the

connection between some user preferences and design

elements. Misaka and Aoyama (2018) use Knowl-

edge Engineering and machine learning for correlat-

ing users based on KANSEI engineering, a method-

ology to transform customer feelings into design as-

Carrera-Rivera, A., Larrinaga, F., Lasa, G. and Martinez-Arellano, G.

UX- for Smart-PSS: Towards a Context-aware Framework.

DOI: 10.5220/0011379700003323

In Proceedings of the 6th International Conference on Computer-Human Interaction Research and Applications (CHIRA 2022), pages 113-120

ISBN: 978-989-758-609-5; ISSN: 2184-3244

113

pects of a new product. However, the studies in this

research area are still limited to user-speciﬁc pref-

erences in the usage stage of Smart PSS. A current

challenge is a way to collect and process user behav-

ior data in real time and deﬁne how it will be used.

Existing research focused primarily on customer data

from surveys or data from online reviews on social

media. Further work on adaptation of the means of

interaction with S-PSS needs to be done in order to

improve the user experience to meet individual user

needs. This paper proposes a framework to provide

an adaptive UX according to user context, consider-

ing the interactions and data from devices and user

proﬁles. The structure of this work is as follows: Sec-

tion 2 starts with the fundamental background related

to UX and Context-Awareness. Section 3 presents

the framework and provides a detailed explanation of

each phase. Section 4 provides a proof of concept of

the framework using a wearable S-PSS. Finally, Sec-

tion 5 presents the description of challenges, future

work and conclusion of this work.

2 BACKGROUND

UX is a result of the internal state of the user, the

characteristics of the designed entity and the context

features where the interaction occurs (Hassenzahl

and Tractinsky, 2006). According to Valencia et al.

(2015), in S-PSS, the UX is characterized by feel-

ings of customer empowerment, the individualization

of services, the sense of ownership, and an individual

and shared experience. Non-engineering approaches

and qualitative tools (i.e., observation, interviews, ex-

perience maps, etc.) are broadly used by UX design-

ers at the very early stage of design for user research,

to understand the context of use and users involved.

Chang et al. (2019) presented a case study for a smart

pillbox directed at the elderly; using behavioural anal-

ysis and interviews they perform a user analysis not

only to capture their speciﬁc requirements but also

to understand the physiological and cognitive dimen-

sions. Similarly, Jia et al. (2021) used behavioural

analysis and experience maps for the development of

a smart rehabilitation assisting device. In both cases,

current users’ problems, popularly denominated ’pain

points’ usually highlight service opportunities.

However, there is a need to exploit the digital ca-

pabilities of S-PSS to generate an adaptive UX with

the collected data available. A data-driven UX is a

particular characteristic of S-PSS where data from

smart products and digital services is easily accessible

(Carrera-Rivera et al., 2022). Furthermore, the ability

to capture user-generated data in real-time should be

an important part of the design process, and it is a way

of identifying and evaluating users’ needs. For in-

stance, methods to collect and process real-time user

behaviour data and deﬁne in what ways it will be used.

Existing research mainly focuses on online reviews

data on social media or customer data from surveys.

Wang et al. (2019) used user ratings and comments

related to smart bicycles service from a website, and

with the use of Natural Language Processing (NLP)

techniques were able to capture implicit requirements

that users had. Similarly, Mourtzis et al. (2018) pre-

sented a framework to evaluate the PSS services us-

ing the feedback received from shop ﬂoor experts and

business customers through social media platforms.

They also considered the information from machines

and manufacturing execution systems to obtain KPI

values. Therefore, user behaviour data from devices

and interactions with service applications are typi-

cally not discussed.

In S-PSS, the delivery of personalized services

can have a great impact on UX to affect the level of

user satisfaction. Multiple approaches beyond PSS

have used adaptive user interfaces as a way to offer

personalization. Reguera-Bakhache et al. (2021) cap-

tured the interaction sequences, consisting of a series

of click events from the interface of a mixing machine

to understand the patterns of interaction from oper-

ators through a clustering analysis, and in this way

reduce the time and interaction sequences of opera-

tors to perform a task. Todi et al. (2021) presented a

system for adaptive menus using the data from user

clicks on menu items and reinforcement learning to

approximate menus to the user’s expertise and inter-

est. However, the use of context can further improve

the personalization and adaptation by capturing the

real needs of the user at a given time.

2.1 Context-awareness

Context is deﬁned by Abowd et al. (1999) as “any

information that can be used to characterize the situa-

tion of an entity. An entity is a person, place, or object

that is considered relevant to the interaction between a

user and an application, including the user and appli-

cations themselves”. In UX, the context of use rep-

resents the users, tasks, equipment (i.e. hardware,

software), and the physical and social environments

in which S-PSS is used (Iso, 2010).

The term “context-aware” was deﬁned by Schilit

and Theimer (1994) as the “ability to discover and

react to changes in the environment”. In general, a

context-aware system uses context to deliver informa-

tion and/or services relevant to a user’s task (Abowd

et al., 1999). Thus, a context-aware system does

CHIRA 2022 - 6th International Conference on Computer-Human Interaction Research and Applications

114

not necessarily imply automatization or real-time pro-

cessing, instead, it refers to the ability to respond to

context. For instance, context can be used to resolve

what services or information need to be presented to

the user. Therefore, context-awareness capability can

be used to personalize the UX by providing services

for a speciﬁc user situation.

In the design of S-PSS, multiple context-aware

approaches have been proposed for the requirement-

elicitation phase. For instance, Wang et al. (2019,

2021) proposed a graph-based context-aware frame-

work for the requirement elicitation process in a data-

driven manner. The framework includes a knowl-

edge management layer that uses domain ontologies

to model the product’s components, services, and

context; and a requirement elicitation layer, which

uses graph algorithms to discover implicit require-

ments that user’s had by analyzing reviews and com-

ments from social media. For UX, context-awareness

has been used to create realistic prototypes. For in-

stance, Seo et al. (2016) proposed a hybrid evaluation

method for smart home services based on VR and AR

prototypes using a UX ontology for smart home ser-

vices and semantic reasoning using the open source

framework Jena. On the other hand, Dou and Qin

(2017) obtained physiological data from several sen-

sors including electrocardiography (ECG) and facial

electromyography (fEMG) from elderly people com-

bined with environmental information, to build a user

mental model to better understand the experience of a

smart TV S-PSS. The authors used clustering analy-

sis as a way to identify patterns and build user proﬁles

according to the context. Such approaches, however,

do not address the need of adaptation in the means of

interaction with the smart products in the usage stage

of S-PSS.

Users’ short term memory is limited to a few

items of information. Cognitive load refers to the

total amount of mental activity on working mem-

ory at an instance in time. If the user interface re-

quires the user to hold more than it can retain, it

will produce a cognitive overload (Tracy and Albers,

2006). Therefore, providing customized services rec-

ommendations address this problem in applications

designed to fully manage or interact with smart prod-

ucts. Context-awareness recommendation systems

have been proved to provide better customization

since context information is very important in improv-

ing recommendation accuracy (Yang, 2018). Tarus

et al. (2018) proposed a context-aware recommender

system for e-learning, using collaborative ﬁltering

(CF) algorithms and learner´s goals and study pat-

terns as sources of context in addition to a traditional

rating of courses. Yang (2018) presented a recom-

Figure 1: Context-aware life cycle.

mender system model, implemented as a website for

movie recommendation. Users are allocated to the

corresponding set of similar users using the Top-N

algorithm. In the user behaviour analysis step, pop-

ular movies are recommended. Then, a matrix of user

scores is established by analyzing the results of user-

registered movies and using time as context of each

score (i.e. season of the year, weekday or weekend).

This section has attempted to provide a brief sum-

mary of the literature. As discussed above, the pro-

posed approaches to UX focus their attention only

on the design stage, before the use phase. As S-PSS

represent a way of continuously striving to meet cus-

tomer needs, adaptation needs to be reﬂected also in

the interaction with smart products. The section be-

low describes the framework proposed in this paper.

3 FRAMEWORK

Perera et al. (2013) deﬁned four general steps in con-

text life cycle as a way to describe the process of de-

veloping context-aware applications (Figure 1). The

ﬁrst step is Context Acquisition, which refers to data

that needs to be acquired from different sources (i.e.

sensors, cyber-physical systems (CPS), databases,

etc). In Context Modelling, the collected data needs

to be represented in a meaningful way. Then, in Con-

text Reasoning, data is processed to provide useful

information and insights (context). Context Dissemi-

nation phase distributes context to consumers, which

can be end-users or other applications. Lastly, Con-

text Monitoring represents the evolution of the de-

sign of S-PSS. It is a stage not broadly represented

and should be considered after Dissemination because

context may change at some point. Systems or appli-

cations have to be able to identify changes and update

the models accordingly.

Following the context-aware life cycle, this paper

proposes a theoretical framework to provide a per-

sonalized UX where services are recommended and

shown according to context when the interaction oc-

curs. Figure 2 presents the overall graphic represen-

tation of the framework. The following subsections

will describe each layer of the framework in detail.

UX- for Smart-PSS: Towards a Context-aware Framework

115

Figure 2: Context-aware framework for UX.

3.1 Context Acquisition

The process of context acquisition will require captur-

ing the data from multiple sources. This process can

be classiﬁed by the source of context and the nature

of the data. Figure 3 presents three main data sources,

Smart Connected Product (SCP), user, and external.

The nature of the data can be dynamic or static. User

context refers to any information related to the user.

User static data might include user personal infor-

mation or user preferences (Liu et al., 2011). The

dynamic information might include: the user’s cur-

rent and historical location, user’s current and histori-

cal activity, user’s current emotion, relationships, etc.

For instance, capturing the data from users while they

interact directly with products or e-services usually

delivered by mobile apps or web applications. SCP

data is acquired from the device through sensors in

smart products that can proportionate relevant infor-

mation of the user when the smart products are used

to monitor the user routine, for instance, wearable de-

vices. Finally, external sources contain dynamic en-

vironment physical information(i.e lighting , noise,

temperature, and humidity level, trafﬁc conditions)

Figure 3: Context data sources.

3.2 Context Modelling

Context modelling allows the representation of the

data acquired in terms of attributes, characteristics,

and relationships with previously speciﬁed context

(Liu et al., 2011). Thus, a static model is necessary

to represent the services of smart devices where each

of them groups low-level services or sub-services re-

lated. A sub-service will represent a very speciﬁc

function of the e-service platform or smart device.

According to (Maleki et al., 2018) in an S-PSS ar-

chitecture, the domain knowledge models function in

an intermediary role to integrate the S-PSS services

with the cyber-physical components. One approach is

the use of domain ontologies. Ontologies are a means

to formally model concepts from a particular domain

into a detailed speciﬁcation of entities with properties

and relations. Domain ontologies deﬁne the vocab-

ulary related to a particular domain (Guarino et al.,

2009). In this case, each service is modelled into a

pattern that relates the product, service and required

information (Maleki et al., 2018) and integrates it with

context-speciﬁc classes and relationships (see Figure

4). In this way, the interactions with the service can

be labelled into multiple categories (i.e. click, like,

close) and then linked to the services to understand

user behaviour for each of the sub-services.

Figure 4: S-PSS Pattern and derived ontology.

3.3 Context Reasoning

Context reasoning is the process of deducting new in-

formation that is useful for a task or user, from multi-

ple context-data sources. The relevance of a recom-

mendable service solution not only depends on the

users’ general preferences but also on their current

situation and their short-term interaction and inter-

ests (Quadrana et al., 2018). Furthermore, the data

from the product itself is a source of context. Col-

laborative Filtering (CF) is a recommendation tech-

nique largely used in e-commerce to provide person-

alized recommendations on items (i.e. books, movies,

CHIRA 2022 - 6th International Conference on Computer-Human Interaction Research and Applications

116

Figure 5: Reasoning process context-aware recommendation systems.

clothes, etc). It uses similarities between users and

items simultaneously to provide recommendations.

CF exploits the ratings provided to products explic-

itly (i.e. starts ratings) to measure the similarity and

provide adequately recommendations (Elahi et al.,

2016). However, is also possible to get implicitly

forms of ratings by obtaining and evaluating the in-

teractions that the user had (i.e. page views, clicks,

purchases, etc), which is the more adequate method

in the case of S-PSS. Each type of interaction could

represent a higher level of engagement with a spe-

ciﬁc service. For example, monitoring events such

as ﬁlter changes and clicks would represent a higher

rating in comparison with a page view event (Chen,

2005; N

nez-Valdez et al., 2018) as seen in Figure 5.

In a context-aware scenario, ratings are modeled as a

function of: users, service as well as context. Hence,

the rating function can be deﬁned in three dimensions

(Tarus et al., 2018): User x Service x Context Rat-

ing, the process will produce as output an ordered

list of recommended services based on the generated

ratings of other users with similar preferences in the

same context (Figure 5). Although collaborative ﬁl-

tering is the most popular recommendation technique,

its major disadvantage is the new user and new item

problems (Barjasteh et al., 2016), referred commonly

as the cold-start problem, which occurs in scenarios

where it is not possible to make reliable recommen-

dations due to an initial lack of ratings (Adomavicius

and Tuzhilin, 2005). Semantic reasoning can be used

to explicitly infer the preference of implicit users, re-

ducing the sparsity of user information, and alleviat-

ing the problems of cold start (Yang, 2018). Consid-

ering the use of ontologies in the modelling stage, se-

mantic reasoning could take advantage of this to infer

new knowledge based on established relationships.

3.4 Context Dissemination and

Monitoring

Context Dissemination is related to the methods to

deliver context to clients or users. The purpose of

this stage is to provide the individualization of the

experience by providing services adequately for each

user situation. Digital services should be adaptive to

user context and easily accessible through digital plat-

forms provided by the service in the form of apps or

websites or the product means of interaction by itself.

Furthermore, the process of context monitoring repre-

sents the evolution of the design of S-PSS. It is impor-

tant to monitor user feedback regarding the services

delivered by the application. Feedback from users can

come through integrated surveys, but also from user

behaviour. For instance, if a user is not happy with

the service that the application is offering, it is likely

they will return to an old state of the application, or do

not interact with the information. This can be taken as

negative feedback and a negative rating for the service

provided. Having discussed each layer, the following

section of this paper addresses an application scenario

to understand the implementation process.

4 APPLICATION SCENARIO

In order to demonstrate the proposed framework, an

activity monitoring scenario is used for a product-

oriented PSS. Tukker (2004) classiﬁed PSS into three

main categories from the business aspect: Product-

oriented, Result-oriented, and Use-oriented. Product-

oriented is related to services that facilitate the sales

of products or add functionality or personalize exist-

ing products. In Result-oriented and Use-oriented,

the ownership usually remains with the provider. Fit-

bit is one of the most popular multi-purpose activ-

ity wearable trackers and smart-watch and represents

a clear example of Product-oriented PSS for end-

consumers (B2C). An important part of the product

UX- for Smart-PSS: Towards a Context-aware Framework

117

is the e-services platform since it represents a means

of interaction with the device and also a source of in-

formation to the user which is accessible through the

mobile app or web application.

Figure 6: Fitbit application scenario.

In the framework, the context acquisition layer

collects the information from the device, which in

this case represents data about the user’s behaviour,

for instance, the number of steps taken by the user

in a day. This translates into information such as the

distance, the duration as well as the intensity of ac-

tivity. User interaction with the e-service platform

will be captured using Matomo

, which is an open-

source platform capable to capture user interaction

(i.e clicks, page views) from a web application into a

Mysql database. Figure 6 represents the context mod-

elling layer. Using an ontology, according to section

3.2, each high-level service provided by the S-PSS

(i.e activity monitoring) has associations with low-

level services that represent speciﬁc options on the e-

service platform. Each interaction that is recorded can

be related to a speciﬁc service and further enriched

with contextual information, for instance, weather,

time and device status. Another signiﬁcant aspect is

that the use of an ontology could mitigate the ’cold

start’ problem when there is a lack of historical inter-

actions. In this scenario, proﬁles were deﬁned using

a dataset from 30 anonymous users (Furberg et al.,

2016) by means of a non-supervised machine learn-

ing technique, using the clusterization algorithm, K-

Means (Figure 7).

Four proﬁles were identiﬁed, in base of their be-

havioural information, and further used to model an

ontology using GraphDB. The ontology can deﬁne

the relationships among users, proﬁles and high-level

and low-level services as shown in Figure 8, User1

has an Active proﬁle and uses trackExercise low-level

service. Using semantic reasoning, it can be inferred

that users with Active proﬁle use the trackExercise

service. In future work, the reasoning layer will use

https://matomo.org/

Figure 7: K-means clustering.

collaborative ﬁltering algorithms to produce an or-

dered list of recommended services. These services

will be delivered through the e-service platform ap-

plication in the form of widgets or personalized no-

tiﬁcations in a web application using React frame-

work. One of the challenges will be to achieve a

way to deliver the adaptive services in several S-PSS

interfaces; software frameworks for application cre-

ation are very diverse. Javascript is the ultimate web

standard with multiple frameworks available (i.e. Re-

act, Vue, AngularJS), which have extended to mobile

apps. But, forms of interactions might vary in indus-

trial (i.e. manufacturing) settings. This highlights the

importance to understand the type of market where

the S-PSS is directed. It is key to validate the frame-

work in both, B2C scenario and also in a Business to

Business scenario (B2B), where the UX can be per-

ceived differently by users, and the means of interac-

tion and feedback recollection can also change.

Figure 8: Ontology query in GraphDB.

Another challenge of a data-driven UX is the

need for users’ personal and behavioural information,

which could vary depending on S-PSS’ application

domains (i.e health & welfare, manufacturing, smart

home, etc.). The services created must help users

make informed decisions about their privacy. Simi-

larly, users must have control over their personal in-

formation (van Ooijen and Vrabec, 2019). Privacy

regulations should be followed (i.e. General Data Pri-

vacy Regulation GDPR) and ensure users’ consent

and awareness of the data collected (Zheng et al.,

2019a).

CHIRA 2022 - 6th International Conference on Computer-Human Interaction Research and Applications

118

5 CONCLUSIONS

S-PSS emerges as a way to provide a market offer

combining both product and service while maintain-

ing a lasting relationship between provider and cus-

tomer. It is important to comprehend the role of UX

to understand the use and expectations of users with

smart products and e-service platforms. Continuously

delivering personalized experiences for users is chal-

lenging. The creation of value with user interac-

tions can allow maintaining the system relevant to the

changing needs. This can be accomplished, by imple-

menting technology in a way that it results in a simple

and intuitive experience. This work described a theo-

retical context-aware framework for the UX of S-PSS,

following the life cycle of context-aware application

and considering the use of context in recommenda-

tion systems and the users interactions. The goal is to

provide a personalized and adaptive e-service S-PSS

platform for each user. An application scenario of a

wearable activity tracker device was presented as a

product-oriented S-PSS that allows to better showcase

the general overview of the framework. The work also

presented a description of the challenges that brings

S-PSS, such as a multi-business perspective to under-

stand users in different scenarios and industries, and

the management of behavioural data and how to han-

dle privacy aspects within the applications. Future

work will provide an implementation of the frame-

work and further analyze each of the stages described

in this paper.

ACKNOWLEDGEMENTS

This project has received funding from the European

Union’s Horizon 2020 research and innovation pro-

gramme under the Marie Sklodowska-Curie Grant

No. 814078.

REFERENCES

Abowd, G. D., Dey, A. K., Brown, P. J., Davies, N., Smith,

M., and Steggles, P. (1999). Towards a better under-

standing of context and context-awareness. In Inter-

national symposium on handheld and ubiquitous com-

puting, pages 304–307. Springer.

Adomavicius, G. and Tuzhilin, A. (2005). Toward the

next generation of recommender systems: A sur-

vey of the state-of-the-art and possible extensions.

IEEE transactions on knowledge and data engineer-

ing, 17(6):734–749.

Barjasteh, I., Forsati, R., Ross, D., Esfahanian, A.-H., and

Radha, H. (2016). Cold-start recommendation with

provable guarantees: A decoupled approach. IEEE

Transactions on Knowledge and Data Engineering,

28(6):1462–1474.

Carrera-Rivera, A., Larrinaga, F., and Lasa, G. (2022).

Context-awareness for the design of smart-product

service systems: Literature review. Computers in In-

dustry, 142:103730.

Chang, D., Gu, Z., Li, F., and Jiang, R. (2019). A user-

centric smart product-service system development ap-

proach: A case study on medication management

for the elderly. Advanced Engineering Informatics,

42:100979.

Chen, A. (2005). Context-aware collaborative ﬁltering sys-

tem: Predicting the user’s preference in the ubiquitous

computing environment. In International Symposium

on Location-and Context-Awareness, pages 244–253.

Springer.

Cong, J.-c., Chen, C.-H., Zheng, P., Li, X., and Wang,

Z. (2020). A holistic relook at engineering de-

sign methodologies for smart product-service sys-

tems development. Journal of Cleaner Production,

272:122737. Publisher: Elsevier BV.

Dou, J. and Qin, J. (2017). Research on User Mental Model

Acquisition Based on Multidimensional Data Collab-

orative Analysis in Product Service System Innova-

tion Process. In International Conference on Engi-

neering Psychology and Cognitive Ergonomics, pages

35–44. Springer.

Elahi, M., Ricci, F., and Rubens, N. (2016). A survey of

active learning in collaborative ﬁltering recommender

systems. Computer Science Review, 20:29–50.

Furberg, R., Brinton, J., Keating, M., and Ortiz, A.

(2016). Crowd-sourced Fitbit datasets 03.12.2016-

05.12.2016.

Guarino, N., Oberle, D., and Staab, S. (2009). What is an

ontology? In Handbook on ontologies, pages 1–17.

Springer.

Hassenzahl, M. and Tractinsky, N. (2006). User experience-

a research agenda. Behaviour & information technol-

ogy, 25(2):91–97. Publisher: Taylor & Francis.

Iso, D. (2010). 9241–210: 2010: ergonomics of human-

system interaction—part 210: human-centred design

for interactive systems (formerly known as 13407).

Switzerland: International Standards Organization.

Jia, G., Zhang, G., Yuan, X., Gu, X., Liu, H., Fan, Z., and

Bu, L. (2021). A synthetical development approach

for rehabilitation assistive smart product–service sys-

tems: A case study. Advanced Engineering Informat-

ics, 48:101310. Publisher: Elsevier.

Liu, W., Li, X., and Huang, D. (2011). A survey on con-

text awareness. In 2011 International Conference on

Computer Science and Service System (CSSS), pages

144–147. IEEE.

Maleki, E., Belkadi, F., Boli, N., Van Der Zwaag, B. J.,

Alexopoulos, K., Koukas, S., Marin-Perianu, M.,

Bernard, A., and Mourtzis, D. (2018). Ontology-

based framework enabling smart product-service sys-

tems: application of sensing systems for machine

health monitoring. IEEE internet of things journal,

5(6):4496–4505. Publisher: IEEE.

UX- for Smart-PSS: Towards a Context-aware Framework

119

Misaka, M. and Aoyama, H. (2018). Development of de-

sign system for crack patterns on cup surface based

on kansei. J. Comput. Des. Eng., 5:435–441.

Mourtzis, D., Fotia, S., Vlachou, E., and Koutoupes, A.

(2018). A Lean PSS design and evaluation framework

supported by KPI monitoring and context sensitivity

tools. The International Journal of Advanced Manu-

facturing Technology, 94(5-8):1623–1637. Publisher:

Springer Nature BV.

nez-Valdez, E. R., Quintana, D., Crespo, R. G., Isasi, P.,

and Herrera-Viedma, E. (2018). A recommender sys-

tem based on implicit feedback for selective dissemi-

nation of ebooks. Information Sciences, 467:87–98.

Perera, C., Zaslavsky, A., Christen, P., and Georgakopoulos,

D. (2013). Context aware computing for the internet

of things: A survey. IEEE communications surveys &

tutorials, 16(1):414–454. Publisher: Ieee.

Pezzotta, G., Cavalieri, S., and Romero, D. (2017). Collab-

orative product-service systems engineering: Towards

an active role of customers and stakeholders in value

co-creation. In 2017 International Conference on En-

gineering, Technology and Innovation (ICE/ITMC),

pages 1247–1254. IEEE.

Quadrana, M., Cremonesi, P., and Jannach, D. (2018).

Sequence-aware recommender systems. ACM Com-

puting Surveys (CSUR), 51(4):1–36.

Reguera-Bakhache, D., Garitano, I., Cernuda, C.,

Uribeetxeberria, R., Zurutuza, U., and Lasa, G.

(2021). An adaptive industrial human-machine inter-

face to optimise operators working performance. In

2021 IEEE/ASME International Conference on Ad-

vanced Intelligent Mechatronics (AIM), pages 1213–

1219. IEEE.

Schilit, B. N. and Theimer, M. M. (1994). Disseminating

active map information to mobile hosts. IEEE net-

work, 8(5):22–32.

Seo, D. W., Kim, H., Kim, J. S., and Lee, J. Y. (2016). Hy-

brid reality-based user experience and evaluation of

a context-aware smart home. Computers in Industry,

76:11–23. Publisher: Elsevier.

Tarus, J. K., Niu, Z., and Kalui, D. (2018). A hybrid recom-

mender system for e-learning based on context aware-

ness and sequential pattern mining. Soft Computing,

22(8):2449–2461.

Todi, K., Bailly, G., Leiva, L., and Oulasvirta, A. (2021).

Adapting user interfaces with model-based reinforce-

ment learning. In Proceedings of the 2021 CHI Con-

ference on Human Factors in Computing Systems,

pages 1–13.

Tracy, J. P. and Albers, M. J. (2006). Measuring cogni-

tive load to test the usability of web sites. In Annual

Conference-society for technical communication, vol-

ume 53, page 256.

Tukker, A. (2004). Eight types of product–service sys-

tem: eight ways to sustainability? Experiences from

SusProNet. Business Strategy and The Environment,

13:246–260.

Valencia, A., Mugge, R., Schoormans, J., and Schifferstein,

H. (2015). The design of smart product-service sys-

tems (PSSs): An exploration of design characteristics.

International Journal of Design, 9(1).

van Ooijen, I. and Vrabec, H. U. (2019). Does the gdpr

enhance consumers’ control over personal data? an

analysis from a behavioural perspective. Journal of

consumer policy, 42(1):91–107.

Wang, Z., Chen, C.-H., Zheng, P., Li, X., and Khoo,

L. P. (2019). A novel data-driven graph-based re-

quirement elicitation framework in the smart product-

service system context. Advanced engineering infor-

matics, 42:100983. Publisher: Elsevier.

Wang, Z., Chen, C.-H., Zheng, P., Li, X., and Khoo,

L. P. (2021). A graph-based context-aware require-

ment elicitation approach in smart product-service

systems. International Journal of Production Re-

search, 59(2):635–651.

Wetter-Edman, K., Sangiorgi, D., Edvardsson, B., Holmlid,

S., Gr

onroos, C., and Mattelm

aki, T. (2014). Design

for value co-creation: Exploring synergies between

design for service and service logic. Service Science,

6(2):106–121. Publisher: INFORMS.

Yang, Q. (2018). A novel recommendation system based

on semantics and context awareness. Computing,

100(8):809–823.

Zheng, P., Liu, Y., Tao, F., Wang, Z., and Chen, C.-H.

(2019a). Smart product-service systems solution de-

sign via hybrid crowd sensing approach. IEEE Access,

7:128463–128473.

Zheng, P., Wang, Z., Chen, C.-H., and Khoo, L. P. (2019b).

A survey of smart product-service systems: Key as-

pects, challenges and future perspectives. Advanced

engineering informatics, 42:100973. Publisher: Else-

vier.

CHIRA 2022 - 6th International Conference on Computer-Human Interaction Research and Applications

120