Towards a core ontology of organisational transformation

2.1. Unified Foundational Ontology (UFO) and OntoUML

UFO is an axiomatic formal theory based on contributions from Formal Ontology in Philosophy, Philosophical Logic, Cognitive Psychology, and Linguistics. UFO has been conceived as a foundation for conceptual modelling. Over the years, it has been successfully employed to support ontological analysis and conceptual clarification in many domains (Guizzardi, 2005; Guizzardi et al., 2008; Botti Benevides et al., 2019; de Oliveira Bringuente et al., 2011).

OntoUML is an extension of UML Class Diagrams whose meta-model has been designed to comply with the ontological distinctions and axiomatisation of UFO (Guizzardi, 2005; Fonseca et al., 2019; Guizzardi et al., 2021; Almeida et al., 2019a).

A recent study showed that UFO is the second-most used foundational ontology in conceptual modelling, with the fastest adoption rate (Verdonck and Gailly, 2016). This study also shows that OntoUML is among the most used languages in ontology-driven conceptual modelling.

UFO is structured into three modules; UFO-A (an Ontology of Endurants) (Guizzardi, 2005), UFO-B (an Ontology of Events) (Botti Benevides et al., 2019), and UFO-C (Guizzardi et al., 2008; de Oliveira Bringuente et al., 2011) (an Ontology of Social and Intentional Entities). UFO-A and UFO-B are directly reflected in the modelling primitives of OntoUML, as well as in the constraints that govern the relations between these primitives. UFO-C has been built on top of UFO-A and UFO-B to systematise social concepts such as plan, action, goal, agent, intentionality, commitment, and appointment, among many others. UFO-C, however, is represented as an OntoUML library with general categories that are later specialised by the corresponding domain-specific notions.

A brief explanation of a selected subset of the ontological distinctions set out by UFO is offered in the remainder of this section. We also show how these distinctions are represented by the modelling primitives of OntoUML (as a UML profile). Following that, we briefly present the categories comprising UFO-A; the two subsequent sections focus on the categories of UFO-B and UFO-C, respectively. For an in-depth discussion, philosophical rationale, formal characterisation and empirical support for these categories, one should refer to (Guizzardi, 2005), (Guizzardi et al., 2008), (Botti Benevides et al., 2019), (Guizzardi et al., 2022), (Verdonck et al., 2019). In this text, we highlight in bold UFO terms the first time they appear and show in «brackets» the corresponding OntoUML stereotypes. One should notice that OntoUML only represents types of entities and (Onto)UML stereotypes predicated over these types, not the entities instantiating these types. As such, when we mention in the sequel that we use the stereotype, e.g., «relator» or «collective» for relators and collectives, what we mean is that these stereotypes predicate over types whose instances are relators and collectives, respectively. Mutatis mutandis, when we mention that we use the relation of «mediation» between relators and their bearers, what we mean is that stereotyped relation is represented between a relator type and another type(s) such that the instances of these types are bound by mediation relations accordingly.

UFO makes a fundamental distinction between Endurants and Events. Endurants are entities that exist in time and can change qualitatively while maintaining their identity. Events, in contrast, are entities that unfold in time and accumulate temporal parts (e.g., a football game, a conversation, a speech). Endurants are further refined into Substantial and Aspects.

Substantials are existentially independent entities. They can be either Functional Complexes (e.g., Mick Jagger, my Car, the Moon) or Collectives («collective») (e.g., a deck of cards, a crowd). Collectives have uniform structures and contain other endurants as members («memberOf» relation).

Unlike substantials, aspects are parasitic entities and can exist only by inhering in other entities (their bearers), i.e., they are existentially dependent entities. Aspects can be Intrinsic Aspects (which depend on one single bearer) or Relational Aspects (which depend on a multitude of bearers, thus connecting them). Intrinsic aspects can be Qualities (e.g., the colour of an apple, the weight of the rock, the inflation of a country) or a Mode (e.g., a belief, an intention, a headache, but also dispositions such as electrical conductivity and magnetic attraction). Enrolment is an example of a relational aspect (also called a Relator in UFO). Qualities and modes are connected to their bearers via an existential dependence relationship of inherence1

Let us first take a domain in reality that is populated by endurants. Entities in this domain will instantiate a number of Kinds, i.e., the genuine fundamental types of endurants exist in this domain. The term ‘kind’ is meant in a strong technical sense – it is a type that captures essential properties of the things it classifies. In other words, the endurants instantiating a kind could not possibly exist without being of that specific kind. Kinds tessellate the possible space of endurants in that domain, i.e., all endurants necessarily belong to exactly one kind. OntoUML uses the stereotype «kind» simpliciter to represent types of substantial whose instances are functional complexes and the stereotype «collective» for those whose instances are collectives. Mode, quality and relator kinds are represented by the homonyms stereotypes (see Figs 2–5). There can, however, be other static subdivisions (or subtypes) of a kind. These are naturally termed Subkinds («subkind»).

Endurant kinds and subkinds represent essential properties of endurants, also termed rigid or static types (Guizzardi, 2005). There are also types that represent contingent or accidental properties of endurants; these are termed anti-rigid types (Guizzardi, 2005). These include Phases («phase»). Phases dynamically classify their instances, for example, in the way that ‘being a puppy’ captures a cluster of contingent intrinsic properties of a dog, or in the way that ‘being a living person’ captures a cluster of contingent intrinsic properties of a person. Living Person and Deceased Person form a partition (termed phase partition), where every person instantiate exactly one of the two phases. Every instance of Living Person is only contingently an instance of Living Person (in the modal sense). For every P such that P is a Living Person, there is a counterfactual situation in which P is not a Living Person, which implies that P is a Deceased Person.

Another example of an anti-rigid type is a Role («role»). Roles are anti-rigid and relationally dependent, for example, in the way that ‘being a husband’ captures a cluster of contingent relational properties of a man participating in a marriage. In other words, the contingent properties represented by phases and roles differ as follows: phases represent properties that are intrinsic to entities (e.g., ‘being a puppy’ is being a dog that is in a particular developmental phase; ‘being a living person’ is being a person who has the intrinsic property of being alive; ‘being an available car’ is being a car that is functional and, hence, can be rented). Roles, on the other hand, represent properties that entities have in a relational context, i.e., contingent relational properties. For example, ‘being a husband’ is to bear a number of commitments and claims towards a spouse within the scope of a marital relationship; ‘being a student’ is to bear a number of properties within the scope of an enrolment relationship with an educational institution.

Kinds, subkinds, phases, and roles are categories of Sortals. In the philosophical literature, a sortal is a type that provides a uniform principle of identity, persistence, and individuation for its instances (Guizzardi, 2005). To put it simply, a sortal is either a kind (e.g., ‘Person’) or a specialisation of a kind (e.g., ‘Student’, ‘Teenager’, ‘Woman’), i.e., it is either a type representing the essence of what things are or a sub-classification applied to the entities that “have that same type of essence”.

Endurants are connected by relators while playing certain “roles”. For instance, people play the role of the spouse in a marital relationship; a person plays the role of president in a presidential mandate. ‘Spouse’ and ‘President’ (but also typically student, teacher, pet) are examples of what was technically termed a role in UFO. It is a relational contingent sortal since this type of role can be played only by entities of a single given kind. There are, however, relational and contingent role-like types that entities of multiple kinds can play. An example is the role of the ‘Customer’, which both people and organisations can play. We call these role-like types that classify entities of multiple kinds Role Mixins («roleMixin»).

Role Mixins are examples of non-sortals (also called dispersive types), i.e., types that classify endurants of multiple kinds. Role Mixins are contingent, relationally dependent non-sortals. There are, however, non-sortals that are rigid and relationally independent. These are termed Categories («category»), which represent essential properties that are shared by endurants of multiple kinds. For example, in a human body («kind»), the heart («kind») is a type of muscle («category»). Muscle is a category as we have muscles in other parts of our bodies.

Events are represented in OntoUML by having their types stereotyped as «event» (Almeida et al., 2019a). The language also includes several relations between events (e.g., parthood), between substantial and events (e.g., participation), and between modes (dispositions) and events (e.g., «manifestation»). Roles (and role mixins) whose relational dependence conditions are defined in terms of participation in events are termed Historical Roles (and Historical Role Mixins) («historicalRole» and «historicalRoleMixin»).

Finally, UFO also considers the existence of Types whose instances are other types (i.e., higher-order types), to form multi-level structures. These classes are represented in OntoUML by classes stereotyped as «Type» (Carvalho et al., 2016; Guizzardi et al., 2015; Carvalho et al., 2017). Following Guizzardi et al. (2015), we take types here to be special sorts of endurants.

UFO-B (Guizzardi et al., 2008; Botti Benevides et al., 2019) explicitly differentiates between events and endurants regarding their relationships with time. Endurants are fully present at whatever instant of time they are present, i.e., they are in time in the sense that, if a substantial S has the property P 1 in a circumstance c 1 , and the property P 2 (possibly incompatible with P 1 ) in a circumstance c 2 , it remains the same substantial S in both situations. For example, John weighs 80 kg in c 1 and 70 kg in c 2 , but he is the same John. Events (sometimes also called perdurants or occurrences) are individuals composed of temporal parts. They unfold in time, accumulating temporal parts. Examples of events are a conversation, a football match, and a symphony performance. Only some of its temporal parts will be present when an event is present. Consequently, events cannot change in the genuine sense since none of their temporal parts retains their identity over time.

For UFO-B (Botti Benevides et al., 2019): (i) events are possible transformations from one Situation to another, i.e., they can change the state of affairs in reality from one (pre-)state to another (pos-)state; (ii) events are ontologically dependent entities as to exist, they existentially depend on their participants; (iii) events can be mereologically complex (following an extensional mereology); (iv) events are manifestations of dispositions; (iv) events relate to each other via temporal relations (Allen Relations) as well as via causal relations. Causation is derived from a mechanism of disposition activation by situations plus having events bringing about situations. In other words, if an event e brings about a situation s, which activates disposition d, which, in turn, is manifested by event e’, then we have that e has caused e’. Moreover, if e causes e’, which causes e”, then e indirectly causes e”. Indirect causation is, thus, a strict partial order relation.

Some authors, most notably (Galton, 2006), make a strong distinction between processes and events. For him: (a) processes are open-ended, whereas events are not, i.e., while events have a beginning and an end, processes go on from moment to moment and, in principle, can continue indefinitely (although extraneous factors can prevent them from doing so). In other words, an event is a piece of history, while a process can be experienced in the present; (b) processes are homeomerous, i.e., if a process goes on over some time interval, then it goes on over each subinterval of that interval; (c) processes (but not events) can be the direct objects of experience. In this respect, processes are almost more like endurants than perdurants: they are present from moment to moment, can be experienced directly, and undergo changes.

In UFO-B, in contrast, all events (atomic or complex, homeomerous or otherwise) are modally fragile entities (i.e., they cannot change) and only exist as past entities (Guizzardi et al., 2016). The term process is sometimes used to refer to complex events, but they are nonetheless entities of the same nature (Guizzardi et al., 2022). The notion of ongoing complex events is detailed in (Guizzardi et al., 2016), and, in a nutshell, ongoing events are modelled as having successive numerically distinct events of the same kind that are proper parts of their successors. A defined description as a constant (e.g., this walk to the station, this fall of the rock) gives the illusion of numerical identity of these different succeeding events and is individuated by a constant focus, namely a background endurant which is either the changing quality at hand (Guarino et al., 2022) (e.g., the vertical position of the falling rock) or the disposition of which the event is a manifestation (Guizzardi et al., 2016) (e.g., an intention to walk to the station).

2.3. An Ontology of Social Entities (UFO-C)

One of the main distinctions made in UFO-C is between Agents and (non-agentive) Objects. Agents (but not objects) are substantials that have intentionality. Intentionality in UFO should be understood in a broader sense than the notion of “intending something”. It refers to the capacity of some properties of certain individuals to refer to possible situations of reality. Thus, “intending something” is a specific type of Intentional Aspect termed Intention. Intentional Aspects can be Social Aspects or Mental Aspects. Social Commitments (i.e., commitments towards other agents) are examples of social aspects. Social Relators are relators composed of social aspects (e.g., commitments and claims).

Intentional Aspects include, besides Intentions, Beliefs (that can be justified by situations in reality) and Desires (expresses a will of an agent towards a state of affairs in reality). Intentions are desired state of affairs for which the agent commits at pursuing (internal commitment, i.e., commitments the agent creates towards itself, self-commitments) towards a Goal (the propositional content of that Intention), i.e., towards bringing about a certain state of affairs in the world. One particular type of Intention is a Closed Intention, which is a commitment to pursue a goal in a specific way, i.e., constrained by a particular type of action Type termed a Plan. Agents can be physical (e.g., a single natural person) or social (e.g., an organisation). There are various types of organisations (for instance, legal and non-legal). This work focuses on legal types of organisations recognised by law as juridical persons, such as corporations, government agencies, non-profit organisations, or international and inter-governmental organisations (such as the United Nations).

Actions are Intentional events, i.e., events that are performed by agents to satisfy their goals. Examples of actions are writing this paper and a business process. Like events, actions can be atomic or complex. A complex action is composed of two or more participations of endurants. These participations, in turn, can be intentional (and therefore themselves actions) or unintentional events. The object participating in an action is called a Resource. A complex action composed of contributions from the actions of different agents is called an Interaction. A dialogue between two agents is an example of an interaction.

Normative Descriptions are types of social objects that, being recognised by social agents, can bring other types of social objects into existence. Normative description can define one or more rules/norms recognized by at least one social agent. Examples of normative descriptions include the Italian Constitution, the University of Twente bylaws, but also a set of directives on performing some actions within an organisation (a description of a plan).

Figure 1 depicts a fragment of UFO-C containing a summary of the categories discussed in this section. As previously discussed, this model is represented as a module in OntoUML.2

OPEN IN VIEWER

Fig. 1.

A UFO-C fragment.

The organisation is the main subject of organisational transformation. Thus, characterising this notion is a central point of our work. We analysed the various concepts of organisation in the literature, according to (a) Guizzardi (2005), (b) Binder and Clegg (2007), (c) Guizzardi et al. (2008), (d) Bottazzi and Ferrario (2009), (e) Almeida and Cardoso (2011), (f) de Oliveira Bringuente et al. (2011), (g) Frank (2011a), Scheer (2012), (h) W3C (2014). Table 1 compares the definitions and highlighted points in common among them.

Organisations can be brought into existence (along with their parts, i.e., organisational units) by one particular type of normative description recognised by an organisation and termed the organisation’s body of norms. The organisation’s body of norms belongs to the so-called constitutive norms (Almeida and Cardoso, 2011). An organisational structure may also include roles, which define tasks, responsibilities, and required skills for a position (Scheer, 2012; W3C, 2014; Frank, 2011b).

In artificial intelligence, there is currently interest in creating artificial agents’ societies (Bottazzi and Ferrario, 2009). This results in an increasing transformation of certain human agent positions into artificial ones in the digital transformation domain. Therefore, it is essential to define what an artificial agent is. An artificial agent is a non-human and non-legal entity to which true legal responsibility cannot be ascribed. An artificial agent may also perform certain actions in an organisation. Organisations are created by physical and social agents but can be sustained by both physical, social and artificial agents (Bottazzi and Ferrario, 2009).

The literature is not consensual about the differences between enterprise and organisation. For Bottazzi and Ferrario (2009), enterprises and organisations are different concepts. Enterprises can be considered a special kind of organisation. For example, a non-profit institute is an organisation, but it is not an enterprise. Nevertheless, for the European Commission (2003) (Binder and Clegg, 2007), an enterprise represents a collection of people organised together into a community or other social, commercial, or political structure. In this paper, we will refer to the enterprise as a type of organisation (Bottazzi and Ferrario, 2009).

For W3C (2014), a formal organisation is recognised in the world at large, with associated rights and responsibilities (as a corporation, charity, government or church), particularly in legal jurisdictions (W3C, 2014). In contrast, an informal organisation is an interconnected social structure that determines how people work together. An informal organisation (e.g., a special interest group) can be formed within a formal organisation.

An organisation can be composed of organisational units, which means that an organisation can comprise other organisations (sub-organisations), which are in some way associated and hierarchically ordered. In some cases, the sub-organisation can be regarded as standalone: a legally recognised business may be part of a larger group or holding company.

The W3C Organisation Ontology (W3C, 2014) represents explicitly the organisational role that the person fulfils (e.g., the W3C ontology models the responsibilities associated with the function). An agent’s situation fulfilling that role within an organisation is then expressed through instances of the membership relationship. The concept of Post represents a position in the organisation that may or may not be currently filled. Posts enable reporting structures and organisation charts to be represented independently of the person holding those posts. Posts can report to other posts. Therefore, a post can exist without someone holding it. In contrast, a membership represents a relationship between a particular person (agent) and the organisation and does not exist unless an agent partakes in the relationship. A post can have an associated role. An organisational structure comprises organisational units and their relationships. The elementary organisational unit, which cannot be further decomposed, is called a position.

In conclusion, as we observe in Table 1, the literature seems to agree that an organisation can act as a social agent. Thus, in this paper, we take that an organisation: is a Social Agent (UFO-C), which comprises a collection of agents organised together into a community (or other social, commercial or political structure, regardless of its legal form). These structures include partnerships or associations regularly engaged in collective activities, with some common purpose or reason for existence that goes beyond those of the set of agents involved in it. An organisation could be a single integrated organisation or a collection of inter-organisational partners.

Nothing is intrinsically valuable (Guarino et al., 2016). Value only exists because people ascribe it to things; thus, value is subjective to a great extent. Customers familiar with technologies might ascribe a high value to a digital business, whilst their grandparents are unlikely to do so. Value depends on the mental aspects of the value-ascribing stakeholders (henceforth value beholders), such as desires, goals, needs, and preferences (Sales et al., 2017). Moreover, for them, value is not a synonym for benefit. Value arises from weighing benefits and sacrifices. For instance, some people consider owning a smartphone of a specific brand (which is likely more expensive) to bring more benefits than the sacrifice of paying for it. Therefore, the value depends on the beholder and his or her mental aspects. However, this does not mean that an object’s intrinsic aspects (or parts) do not influence the value people ascribe to it.

According to Guarino et al. (2016), the analysis of value, value ascription mechanisms, and their motivating and influencing factors is helpful for decision-making processes. Several approaches have been developed in the value modelling literature, such as the Resource, Event, Agent (REA) Ontology (McCarthy, 1982), e3value (Gordijn and Akkermans, 2001; Gordijn, 2004), and the Value Delivery Modelling Language (VDML) (OMG, 2018). Guarino et al. (2016) present and discuss the Value Ascription Ontology (VAO), aiming at elaborating on the nature of the value ascription relationship. Sales et al. (2017) built upon the Value Ascription Ontology (VAO) to propose a theory of value propositions grounded in UFO. This theory elaborates on the nature of (economic) value, which “things” can have value, and how the process of ascribing value works. In particular, the authors adopt the interpretation of value as use value (or value-in-use), henceforth just ‘value’, as opposed to exchange (or market) value, given the former’s role in formulating value propositions. In this sense, the value of a thing is determined by how well its affordances match the goals/needs of a given agent in a given context.

An organisational transformation aims at value creation for the business. Although its focus is value creation, at the same time, the transformation does come with costs and risks, sometimes unforeseen. Therefore, it is essential to understand the opportunities and potential challenges surrounding its value creation. Since, in an organisational transformation, the nature of the value ascription (for example, the customer relationship) is more important than the value exchange between subjects, as value ascription concept considers “what” value is and “why” something is valuable, while the concept of value exchange between subjects disregards this understanding, we look to clarify the concepts involved. To that end, in our work, we reuse the concepts related to value and risk from the Common Ontology of Value and Risk (COVER) (Sales et al., 2017,2018). Moreover, we agree with (Guarino et al., 2016) on the importance of understanding “what” value is and “why” something is valuable.

Organisations need well-designed business strategies to gain a sustainable competitive advantage. One crucial aspect informing the design of business strategies are value propositions, which are types of value ascriptions. Value ascription is a judgement relationship between an agent (the value beholder) and a value bearer, which the beholder judges as having value for someone (the value beneficiary). A value ascription is an aggregation of “smaller” judgments, i.e., the value ascription components. Each component focuses on the beneficiary’s experience from the perspective of one of its mental aspects (including beliefs).

Value propositions are the answer to why customers engage in a business relationship with one organisation and not another – customers choose (Product and Service) offerings which they perceive to offer a superior value proposition. It may be beneficial for an organisation to redesign itself and its offerings to maximise the value created for its customers. Offerings describe how a product, service or feature create value for a target market segment. A business is segmented into market segments to better design offers to its customers and bring value to the business (Sales et al., 2017).

For Sales et al. (2018), value and risk are intrinsically connected. COVER shows this duality between value and risk, considering them both to be the same type of phenomena but with opposite polarity. Value is a composition of benefits (which emerge from goal satisfaction) and sacrifices (which emerge from goal dissatisfaction). Risk is one type of sacrifice, as a value reducing factor. This suggests the process of ascribing value is strongly related to the process of assessing risk.

Given this strong connection between these two concepts, COVER generalises the notions of value ascription and risk assessment into a type of relator that involves: (i) an agent who is responsible for the judgment, deemed the value and risk assessor, respectively; and (ii) the target of the judgment, that could be either an object or an event. According to COVER, the value/risk ascribed to an object is always derived from the value/risk ascribed to events (or experiences) “enabled” by their dispositions, regardless of whether these events are Intentional or not, and whether they affect one’s goals positively or negatively.

Representing value and risk as qualities means that they can also be measured according to a given scale, such as a simple discrete scale (e.g., ⟨Low, Medium, High⟩) or a continuous scale (e.g., from 0 to 100) (Sales et al., 2018).

2.6. Digital Transformation and Innovation Business Process Management Ontology (DT&I-BPM-Onto)

Digital transformation is defined as the combined effects of several digital innovations bringing about novel actors, structures, practices, values and beliefs that change, threaten, replace or complement existing organisations, ecosystems and industries (Loebbecke and Picot, 2015; Mangematin et al., 2014). Most of the new business forms related to digital transformation also include incorporating new business processes. The digital transformation through robotic process automation at Deutsche Telekom (Schmitz et al., 2019), and the well-known Uber case, are just two of the many examples where processes and technologies are side by side in promoting transformation.

In order to characterise processes and technologies side by side promoting transformation, Bogea Gomes et al. (2020a) proposed a Digital Transformation and Innovation Business Process Management Ontology, DT&I-BPM-Onto. DT&I-BPM-Onto provides a conceptual basis for business organisations to learn about Business Process Management (BPM) and digital transformation concepts and relationships, contributing to leveraging BPM in the digital era, exploring its relationship, and investigating to what extent BPM affects and is affected by digital transformation. In the following paragraphs, the terms written in versalete correspond to DT&I-BPM-Onto entities. Figure 2 presents a DT&I-BPM-Onto fragment with the main entities from BPM and Digital Transformation and Innovation.

OPEN IN VIEWER

Fig. 2.

DT&I-BPM-Onto fragment (from Bogea Gomes et al., 2020a).

DT&I-BPM-Onto specialises the types of digital transformation and innovation (DT&I) into (business model, product, business process). One or more DT&I agents perform a digital transformation and innovation (DT&I). Individual, organisation, society and software agents are types of DT&I agent. Customer, adopter, user, investor and supplier are roles played by DT&I agent. DT&I-BPM-Onto is motivated by DT&I drivers that lead an organisation to carry out a digital transformation and innovation.

A business has one or more business goals, an organisation’s short-term or long-term target that one or more DT&I outcomes achieve. A DT&I outcome is a result of one or more DT&I. DT&I-BPM-Onto explores the digital transformation and innovation (DT&I) as an explicit part of a business goal that through business goal strategy and business goal strategy offering turns ideas into value, resulting in one or more DT&I outcomes. The term DT&I object embodies what the digital transformation and innovation will transform. In innovation, with a high degree of newness, the DT&I object is a revolutionary DT&I object, which may be a new business model, product, business process (not existing in the organisation or even in the world), or still, it may exist but requires some significant change on the part of adopters. This is one of the points we had to review when grounding the ontology in UFO. We realised that what we named object needed to be re-evaluated and re-modelled. In COOT, this is related to the notion of transformable entity.

The transformed business process operationalises the transformed Business that can be embedded in one or more business environments, which in turn influences one or more business organisations. The DT&I-BPM-Onto model is organised into three groups of concepts that are independent from but complementary to each other: BPM, Digital Transformation and Innovation and Digital Technology. DT&I-BPM-Onto was developed with the aim of being a digital transformation ontology. However, while analysing the concepts related to organisational transformation, it became clear that it would be necessary to review some of the ontological choices of DT&I-BPM-Onto. In this sense, as described in Section 4, some of the DT&I-BPM-Onto classes have already been revised in COOT.

A new version of DT&I-BPM-Onto (henceforth simply called DTO) grounded in UFO will be developed as part of future work reusing the analysis and representation put forth in this paper.

3.1. A Systematic Approach for Building Ontologies (SABiO)

To develop the ontology presented in this paper (COOT), we followed the Systematic Approach for Building Ontologies (SABiO) (de Almeida Falbo, 2014), which supports the development of domain and core ontologies, as well as reference and operational ontologies. Ontologies encode knowledge and make them reusable at various levels. They are used by people, databases, and applications that need to share information. Ontologies “include a range of models of varying degrees of semantic richness and complexity” (Obrst et al., 2002). This paper takes a human-oriented approach, resulting in a conceptual model intended for communicating knowledge between humans.

SABiO prescribes an ontology development process comprising five phases: (1) Identification of Purpose and Elicitation of Requirements; (2) Ontology Capture and Formalisation; (3) Design; (4) Implementation; and (5) Testing. However, since COOT is a reference ontology, we are using SABiO’s reference ontology development process that involves only three phases: (1) Identifying Purpose and Eliciting Requirements, (2) Capturing and Formalising Ontology and (3) Testing the Ontology. The Design and Implementation phases are specific to the operational ontology development process. Moreover, SABiO considers five main supporting processes spanning the development process: Knowledge Acquisition, Reuse, Documentation, Configuration Management and Evaluation.

SABiO’s Identifying Purpose and Eliciting Requirements phase comprises four activities that occur iteratively in the following order: Identifying Purpose and Intended Uses, Eliciting Requirements, Identifying Competency Questions and Modularising the Ontology. Ontology requirements can be divided into functional and non-functional requirements. Functional requirements refer to the knowledge (i.e., content) to be represented by the ontology. They can be stated as competency questions, i.e., questions that the ontology should be able to answer. This determines what is and what is not relevant to the ontology. The competency questions establish the ontology’s scope and provide a means to evaluate it. The non-functional requirements refer to characteristics, qualities and general aspects that are not directly related to the ontology’s content. A literature review was conducted in previous research to identify and organise the relevant concepts and relations with the aim of eliciting the requirements for the proposed model. These requirements are presented in the following Section 3.2.

In the phase of Capturing and Formalising Ontology, SABiO suggests that concepts and relations in a reference domain ontology should be identified, organised and analysed in light of a foundational ontology. For building reference domain ontologies, highly-expressive languages should be used. An example of an ontology representation language suitable for reference ontologies is the UFO-based language OntoUML. To develop COOT, we use UFO and OntoUML.

Once the reference ontologies are developed, they need to be evaluated (de Almeida Falbo, 2014; Guarino, 2004; Brank et al., 2005; Poveda-Villalón et al., 2012).

According to Brank et al. (2005), various ontology evaluation approaches have been considered in the literature for different types and purposes of ontologies. The authors present the following categories of evaluation: (b1) those based on comparing the ontology to a gold standard ontology3

Suarez-Figueroa et al. (2013) also consider different approaches to help developers during the ontology evaluation process: (s1) comparison of the ontology to a “gold standard”, (s2) detection of common errors from catalogues in the ontology, (s3) use of dimensions and criteria for describing the quality and goodness of the ontology, (s4) use of the ontology in an application and evaluation of the results, (s5) comparison of the ontology with a data source from the domain to be covered, and (s6) evaluation by experts who check the ontology against the requirements. In addition, according to those authors, ontology evaluation can be supported by natural language techniques.

Walisadeera et al. (2016) argue that there is no agreed method or approach to evaluate an ontology in the current literature. The choice of a suitable approach depends on the purpose of validation or evaluation, the application in which the ontology is to be used, and on what aspect of the ontology we are trying to validate or evaluate.

Here, in line with SABiO’s Testing the Ontology phase, a COOT test case was designed, which involves two main evaluation processes: verification and validation. Verification aims to ensure that the ontology was built correctly, i.e., the artefact meets the competency specifications imposed at the outset. Validation, however, aims to ensure that the right ontology was built, i.e., the ontology fulfils its specific intended purpose. For verification, SABiO suggests checking whether the ontology meets its requirements during the Ontology Capture and Formalisation phase. This can be done by indicating which ontology elements (concepts and relations) are able to answer each competency question formulated. The methodology suggests that potential users evaluate the suitability of the ontology for their intended uses to demonstrate its ability to represent real world situations. In our case, in line with (Duarte et al., 2016), we used real data from a car rental company to instantiate our ontology.

We conducted a literature review to identify research works on the conceptualisation and modelling of the organisational transformation phenomena. We collected, classified, and synthesised the literature findings. In particular, we are interested in studies that investigate the differences between the terms organisational transformation and organisational change; the concepts, constructs, conceptualisations and definitions of these two terms, and the conceptual models, ontologies, strategies and theories used to represent or explain them.

The objective of the literature review was to use its outcome to define the knowledge to be represented by COOT (its functional requirements according to SABiO) and respond to the competence questions that the ontology should answer. The literature review was conducted based on the guidelines provided by (Kitchenham et al., 2009; Boell and Cecez-Kecmanovic, 2015). We selected the following sources for our search: ACM Digital Library, AISeL, EBSCO (includes IEEE Xplore Digital Library, Science Direct and Emerald Insight), Scopus, Google Scholar, and Springer Link.

We acknowledge that all systematic studies are limited by the search string being used. We included in our search string all variations of the terms related to organisational transformation and organisational change. As always, some works might be left out of the analysis, albeit potentially relevant to the domain at hand. The main findings from the literature review are presented below.

Organisations’ environments have been changing rapidly for them to survive and prosper (Binder and Clegg, 2007). Organisations themselves are increasingly executing complex transformations, such as mergers and splits, chain redesign, sharing and sourcing, and the rationalisation of products, processes and systems (Op’t Land et al., 2009; Op’t Land and Dietz, 2012). Thus, organisations require a broader definition to incorporate, as a single autonomous legal entity, the collaborative relationships outside its boundaries demanded by globalisation, outsourcing and virtualisation.

The two central concepts of COOT are the organisation and its transformation. In the following text, the terms written in versalete highlight COOT classes. The ontology COOT is presented diagrammatically in a set of modules, namely The Organisation Concepts on the Organisational Transformation (Section 4.1.1 and Fig. 3), Motivation for Transformation (Section 4.1.2 and Fig. 4), The Transformation Initiative (Section 4.1.3 and Fig. 5), and The Transformation Event (Section 4.1.4 and Fig. 6). As previously mentioned, COOT is grounded in UFO and its fragments: UFO-A (Guizzardi, 2005), UFO-B (Botti Benevides et al., 2019; Almeida et al., 2019a), and UFO-C (Guizzardi et al., 2008; de Oliveira Bringuente et al., 2011). The categories of UFO-A and UFO-B are directly reused by building the models in OntoUML; the categories of UFO-C are reused by extending the model depicted in Fig. 1. Moreover, in line with SABiO, COOT reuses COVER (Common Ontology of Value and Risk) (Sales et al., 2017,2018) and defines the concept of organisation based on the W3C Organisation Ontology (W3C, 2014), as well as some elements from the DOLCE Organisation Ontology ((Bottazzi and Ferrario, 2009) and (Almeida and Cardoso, 2011)).

OPEN IN VIEWER

Fig. 3.

The organisation, its parts, members, intentions, and assets.

4.1.2. Motivation for transformation

Figure 4 shows the COOT module dealing with organisational transformation’s motivational aspects. Central to this model is a generalisation of the notions of Value Ascription and Risk Assessment from COVER (Sales et al., 2018) into the notion of prospect7

⁷

We thank João Paulo Almeida for suggesting the use of this term here.

assessment. Prospect means an apparent probability of advancement, success, and an outlook for the future. A prospect assessment is a relator binding a prospect beholder (an agent, typically the organisation itself, or one of its subparts or an agent member of the organisation), a prospect bearer, Value Entity or a Value Experience (e.g., a service delivery),8

⁸

As discussed in (Sales et al., 2018), a value experience is an event in the sense of (Guarino, 2017). These non-classical kinds of events are represented here with the stereotype «variable event».

and a prospect beneficiary. prospect assessments can be value ascriptions (to given value bearers) or risk assessments (of a risk subject) (Sales et al., 2018). One particular type of value ascription is a value proposition (Sales et al., 2018) connected with a market segment, which, in turn, is a collective of value beneficiaries (here also generalised into the notion of prospect beneficiary). The notion of value entity here is a generalisation of the COVER notion of value object to include cross-categorical types such as organisational asset.

OPEN IN VIEWER

Fig. 4.

Motivation for transformation.

A prospect assessment is created by a prospect evaluation. This event also leads to the formation of organisation beliefs (sort of UFO-C beliefs), which inhere in organisations. As we shall see, these beliefs account for changes to be promoted in organisational assets, including business offerings. business offerings are specialisations of offering relationships (in the COVER sense) and bind an organisation to an eligible market segment (a subset of a COVER market segment). Sales et al. (2017) argue that value propositions and offerings are directly related, as at the business level value propositions provide an abstract idea of the value offered by the organisation through all of its offerings for a specific market segment, measuring benefits and costs with coarse granularity. While at the offering level, value propositions regarding benefits and costs are very accurate. For them, an offering is a promise with conditional content made by an agent (the offeror) directed to a group of agents that form an eligible market segment. A value proposition (a type of value ascription) describes what an organisation delivers to its customers. In other words, it is the ultimate answer to why customers choose to hire their services or buy their products.

Organisations compete for various reasons (Sales et al., 2018), e.g., to offer better products at lower prices. Often, organisations ultimately compete for customers (Watson, 2002). In competitive markets, after evaluating, assessing, and addressing potentially favourable and unfavourable prospects, organisations carefully design Value Propositions that are aligned with business strategies.

4.1.3. The transformation initiative

The COOT module addressing a transformation initiative is depicted in Fig. 5. Here we have organisation beliefs as the basis for forming particular business intentions. A business intention (i.e., a closed commitment of the organisation restricted to a plan) may or may not become (i.e., move on to a phase of) a transformation initiative.

OPEN IN VIEWER

Fig. 5.

The transformation initiative.

A transformation initiative is a business intention for which agents (transformation initiative participants) have been allocated via social relators termed transformation initiative participant allocation. A transformation initiative participant allocation is thus a bundle of commitments and claims (vis-à-vis UFO-C) binding an agent to play certain organisational roles within the scope of a transformation initiative. The beliefs underpinning the formation of a transformation initiative are termed drivers, i.e., drivers of change. Drivers are organisational responses to prospect assessments.

The organisation brings about an organisational transformation either to mitigate risks, e.g., by eliminating threats or vulnerabilities (Sales et al., 2018) or seizing opportunities envisaged in the assessment of Value Experiences (Sales et al., 2017).

An organisational transformation is a process in the sense discussed in (Guizzardi et al., 2016,2022), i.e., it is a (typically mereologically complex) event manifesting a transformation initiative (a complex closed intention constrained by a transformation plan). That intention is what individuates the temporal parts of that event.

OPEN IN VIEWER

Fig. 6.

The transformation event.