Engineering the Organization from the Bottom Up
Marielba Zacarias
, Rodrigo Magalhaes
and José Tribolet
University of Algarve, Faro, Portugal
Department of Information Systems and Computer Science
Instituto Superior Técnico, Technical University of Lisbon, Portugal
Organizational Engineering Center, INESC Inovação, Lisbon, Portugal
Kuwait Maastricht Business School, Kuwait
Abstract. The paper asserts that the process of emergence which constitutes the
cornerstone of contemporary sociological thought on organization lies also at the
root of the process of organizational engineering. Furthermore it proposes that if
the study of HRIS has crucial organizational implications, then the study of HRIS
should encompass also engineering and modelling considerations. An
organizational modelling framework is put forward which contains the following
propositions, also applicable to the design of HRIS from an integrated
perspective: (1) Enhanced traceability of organizational agents, (2) Situated
enterprise modelling, (3) Model acquisition from action repositories, (4)
Capturing and modelling work practices, (5) Aligning design and execution.
1 Introduction
An agenda for Human Resources Information Systems (HRIS) research is put forward
by [1] arguing from an integrative perspective and stressing that research in
organization and information systems cannot be separated. They argue that the
integration of HRISs within organization can be seen as an intricate web of many causes
and many consequences and that HRIS cannot be studied separate from the organizational
context where they are interwoven. Hence when researching HRIS from an integrated
perspective, it is crucial that the researcher approaches the topic from an appropriate
ontological point of view.
Social emergence is the ontological point of view defended by those authors. [2]
explains that the emergence paradigm research “focuses on the micro-interactional
mechanisms by which shared social phenomena emerge and on how those emergents
constrain those mechanisms” (p. 213). In this paper we discuss an issue which is
relevant to HRIS, i.e. the problem of modelling individual-level behaviour in the
context of broader organizational action. Hence, it is important that the problem under
review is placed within an ontological framework of the organizational phenomenon.
Ontology and methodology are two sides of the same coin, meaning that the
methodology used to research a particular phenomenon will depend entirely on the
ontological perspective that one holds.
Zacarias M., Magalhaes R. and Tribolet J. (2008).
Engineering the Organization from the Bottom Up.
In Proceedings of the 2nd International Workshop on Human Resource Information Systems, pages 97-110
DOI: 10.5220/0001744100970110
Whereas current modelling efforts are mostly directed at organizational
perspectives, little attention has been paid to individual or inter-personal perspectives.
Several approaches to modelling organization strategy, processes and resources have
been developed. However, models for individual or inter-personal levels are scarce
and have typically, different purposes. Research is needed to address the modelling of
individual and interpersonal behaviours and the definition of proper ways of linking
these behaviours with perspectives of higher organizational levels. More specifically,
research is needed to raise awareness and to illustrate the benefits of aligning
individuals and the organization. The aims of such modelling are as follows:
Enabling the organization to capture and visualize different concerns of
individual behaviour.
Enabling individuals to understand the relationship of their daily actions
with organizational resources and activities.
Facilitating the analysis, discussion and (re)design of individual and
inter-personal work.
Organizational modelling and organizational engineering are of interest to HRIS due
to the closeness of this category of information system and all organizational
phenomenon. If organizations are defined essentially as groups of people working for
a common goal, then it is clear that any information system dealing with human
resources will tightly interwoven with the organization itself. There are many schools
of thought in organizational modelling but in this paper we are particularly interested
in a school of thought guided by the following characteristics: (1) Based on the actual
activity of organizational agents, (2) Situated in the actual contexts where agents find
themselves, (3) The model is acquired from action repositories, (4) The actual work
practices are captured and modelled, (5) The design and execution of work practices
are not considered in isolation but in an aligned fashion.
The school of thought in organizational modelling which is followed in this paper
considers organizational phenomena as being emergent in nature. Hence, we begin the
article by putting forward a model of emergent organization. The model is inspired on
the evolutionary logic of autopoiesis which explains the construction of social groups
starting from their biological origins and on the hierarchy of self-referential social
systems put forward in [3]. Each level exhibits to the same autopoietic characteristics
of operational closure and self-referentiality and represents a level of sensemaking at
which the organization can be analysed or diagnosed. The remainder of the paper is
devoted to an exposition on the proposed modelling approach.
2 The Emergent Organization
[3] argues that although autopoiesis cannot be transferred as a whole to social theory,
there is one key principle of autopoiesis which can - the principle of organization
closure. Such argument is based on the assumption that throughout the entire
hierarchy of systems, as proposed by [4], all the systems’ levels exhibit characteristics
of organizational closure. As we have seen above, for autopoiesis the main guideline
for the characterization of living, autonomous systems is not a set of inputs and
outputs, but the nature of their internal coherence, which arise out of their
interconnectedness [5]. In turn, organizational closure “requires some form of self-
reference, whether material, linguistic or social, rather than the more specific process
of self-production” [3, p. 111]. Thus, it is suggested that organizational closure and
self-referentiality are criteria which unequivocabily define social systems.
There are many simple examples of organizational closure and self-referentiality in
every-day life. Conversations are one case in point. In order to maintain its internal
coherence, a conversation between two persons has to be self-referential, meaning
that it must anchored on statements already made and for the conversation to remain
meaningful it must build on past knowledge. Our own perception of events around us
is also self-referential. An example comes from Gestalt theory in psychology and
concerns the phenomenon of apparent movement. When the light in one place is
turned off and the light in another place is immediately turned on, we experience the
perception of light movement. This illusion is the basis of the apparent movement of
neon advertising signs. The observer does not see two lights going on or off and she
immediately infers that something is moving. The immediate perception, on the basis
of past knowledge, is one of movement and it is only by careful analysis that the
observer realizes that there was no physical movement [6].
In Table 1 it is explained how social systems evolve from the level of the
individual to the level of society, consistently maintaining the attributes of
organizational closure and self-referentiality. Starting from the non-social individual,
enacted cognition theory [7] posits that knowledge of the world is formed through the
establishment of enduring relationships between the movement of the body and the
changes in the neuronal activity of the brain. In the words of [8] “to know is to
evaluate through our living, in a creative circularity" (p. 260).
The next stage is the stage where the first inter-personal bonds are created. In order
for the non-social individual to become a social individual the first and crucial
ingredient is communication, the most fundamental social category. As defined by
Luhmann, communication is “the reciprocal interaction between two individuals” [3,
p. 116]. Whereas actions may not be inherently social, communication is always
social and for action to be classified as social there must be communication involved.
Furthermore, communication generates understanding, meaning, emotions and
behaviour, the bases for the formation of bonds between people. “Double
contingency” is the basic mechanism behind the creation of such bonds.
“Double contingency” is an expression coined by Luhmann [3] to explain the
situation that everybody faces in interpersonal interactions of not knowing what the
other person knows or thinks. Given that knowledge is personal and self-referential,
when we speak or when we listen our interpretation of what we said or of what we
heard is always subjective and we are permanently engaged in an ongoing effort to
“guess” what the other person’s expectations are. Thus, double contingency can be
summed up in the following sentence: “I will do what (I think) you expect of me if
you do what (you think) I expect of you”. Still according to Luhmann, it is the
resolution of this daily conundrum that leads to the establishment of an emergent
order of regular patterns of behaviour known as social structure.
Table 1. Emergent levels of self-referential (social) systems.
Type of
Mode of
closure or
Interaction generates
society and society
structures interaction
Closed networks of
communication bound
by structural rules
reproduced through
social interaction
within groups
Structural coupling to
a behavioural domain
in terms of meaning,
legitimation and
Conversations Enduring social or
cultural practices
The social
between people
Expectation of
other’s behaviour in
terms of meaning,
emotion and
Creation of inter-
personal bonds
Body, action
and nervous
Neuronal and bodily
Enactive or
Source: Modified from Mingers (2001)
The explanation regarding the level of social networks and its evolution to the level
above - society/organization - rely also on the social theory developed by Anthony
Giddens. [9] makes an important contribution to an understanding of how social
systems are formed and how reality is socially constructed. For that author, the
evolution of society is radically different from the evolution of living organisms in
that society is a human production. Giddens’ central proposition - structuration theory
- provides a conceptual basis for explaining how social systems are formed through
communication, with new meanings and new words being generated through a
continuous process of narrative making by social actors.
Social boundaries, social norms, and the emerging social practices transcend the
individual and remain even after the individuals have departed. Particular members
may join or leave but the social organization carries on. This is true of small groups,
such as families, micro-communities or sub-cultures in the workplace but it is also
true of larger groups such as clubs, associations, firms, armies or nations. The
transcendental or extra-subjective properties of social organization are the same at
both the level of social networks and of society/organization.
3 Individual-Organizational Alignment
If we now turn to the individual level (non-social individual) of organizational
emergence discussed above, there are some key questions that individuals may
legitimately ask about their organizations. Questions such as:
• who am I in this organization, that is, which roles do I play, what work do I
• how, when, where or why is work accomplished here?
Answers to these questions will clearly help in the effort towards the next level of
emergence, i.e. the social individual. Such evolution is done through a process known
as sensemaking [10, 11]. Sensemaking is the basic mechanism which allows the
process of socialization to evolve in human beings. For sensemaking to ensue a
degree of “alignment” between the individual and its environment (i.e. the
organization) is required.
Individual-organization alignment refers firstly, to the capacity of answering the
questions related to the individuals and organizations as a whole. Second, it refers to
achieving an acceptable level of coherence between individual and organizational
answers. For example, the roles a given individual play in the organization should be
consistent with the roles the individual thinks he plays; or the particular ways that
individuals have of accomplishing activities should be in line with organization’s
processes and goals. Lastly, achieving this coherence should require a reasonable
amount of effort. On one side, the organizations should be tooled with methodologies
and technologies to retrieve the proper information about individuals. On the other
side, individuals should be able to relate their work to organizational processes and
An individual-organizational alignment means that the organization and its
individuals work together to organize the flow of processes and resources such that
they both address individual requirements and the strategic objectives of the
organization [12]. An adequate alignment level between individuals and organizations
enables innovation and, consequently, a competitive and sustainable advantage of the
organization. Enterprise models need to reflect this interaction between individual and
organizational views.
Agents are typically defined as active component parts of organizations.
Individuals, groups and enterprises are all organizational agents that correspond to
different levels of organizational behaviour. Hence, aligning individuals and
organizations entails aligning organizational agents of different levels. In practice,
achieving this kind of alignment has proved to be neither straightforward, nor easy.
Despite the existence of several IS/IT tools for this purpose, this alignment is partial,
frequently inconsistent or outdated.
Organizational agents are not only active. They are also adaptive and
interdependent entities. Hence, they both change in time. Moreover, changes on
individual agents trigger change in collective agents and vice-versa. Therefore, the
alignment of organizational agents need to be addressed as a dynamic and continuous
process. Enhancing organizational modelling to facilitate the alignment between
individuals and organizations as defined in this section, and illustrating its benefits,
are the main motivations of the present research.
4 The Problem Statement
It is submitted, firstly, that the emergence of the organizational phenomenon depends
to a large degree upon the alignment between the individual and the organization.
Secondly, that such alignment cannot be taken for granted; rather, it requires
conscious, systematic and continuous efforts. Thirdly, that the alignment of the
individual and the organization can be facilitated by (1) the development of a semi-
formal models of agent behaviour at different organizational levels and (2) methods
and tools to build, update and analyze the representations based on those models. It is
believed that on-going IT developments, particularly in semantic technologies, data
mining, and enterprise applications, should be explored in enhancing this type of
5 Organizational Modelling
Models are abstractions of real life systems [13]. The use of abstractions has several
benefits [14]. First, it provides an increased ability of processing more information
and/or to process information more quickly. Second, it facilitates the communication
of knowledge to others. Third, abstractions give enormous powers of thought.
Therefore, models simplify and enhance the ability to reason about the system
modelled by omitting certain aspects according to specific purposes and points of
Modelling is present in almost every discipline. One overlapping work area of both
organization science and IS communities is Enterprise –or Organization- Modelling
(EM). Organizations communicate, document and understand their activity through
models [15]. In organization sciences, models are visual representations of given
theories, described in terms of concepts and their relationships [14]. Organization
theorists use them to make abstractions more tangibles. In this field, the main goal of
models is to provide ways of thinking about the organization and to produce
management principles based on these ways of thinking. These models have a high
level of abstraction and are described in natural language. Thus, they are limited to
human use and may lead to different interpretations.
EM has also been addressed by two fields related to computer sciences: IS and
Artificial Intelligence (AI). In these fields, it has been mainly used as communication
tools to facilitate the design and implementation of business applications [16]. Despite
their differences, the frameworks developed in these fields share some characteristics.
First, they allow representing different concerns of enterprise in terms of several
perspectives, dimensions or architectural viewpoints. Second, these perspectives are
inter-related, that is, means of relating concepts from different perspectives are
provided. Third, enterprise models are described with semi-formal or formal
languages and most of them enable graphical representations.
Current EM frameworks are restricted to concerns relevant to system stakeholders.
Moreover, these models are not consistent with the contemporary paradigm of
organizations, since they are based on static, mechanistic and deterministic views of
the phenomenon. They are also based on an objective position of reality, that is,
organizational representations offer an ’aerial’ view, are assumed to be unique and
shared by all members of the organization. Another limitation of current EM
approaches stems from the model acquisition process. Several frameworks provide
means to capture the data required to build the models. These means have varying
levels of detail and support among the different approaches. In general terms, the
acquisition process is mostly manual, and supported by data collection techniques
including interviews, surveys, text/document analysis, among others. This type of
acquisition requires effort and is time-consuming, thus hinders updating
representations to reflect organizational changes, as soon as they take place.
Regarding model acquisitions, the development of information technologies (IT)
has increased dramatically the number and frequency of computer mediated
interactions among individuals. The value of emerging IT is not restricted to
supporting daily operations. Footprints of these interactions can be found within the
repositories of all these applications. Enterprise applications also provide analytical
power, with tools allowing the discovery of hidden patterns in data. Several
frameworks have been proposed to use these applications to enhance and accelerate
sensing and reacting capabilities of organizations. The development of semantic
technologies allows the extracting relevant patterns from non-structured computer-
mediated interactions. Hence, these technologies can be used in further enhancing the
analytical power of enterprises. The combined use of all these technologies looks
promising in facilitating the acquisition and update of enterprise representations from
actual actions and interactions among organizational members.
It is important to note however, that an essential pre-requisite for the successful use
of all these technologies entails overcoming the former limitation. This means
developing models addressing different concerns of organizations and its human
resources. The definition of these models need necessarily to be supported by
exploratory research works reflecting on the nature of organizations, its human
resources, and the critical questions that need to be answered.
6 A Proposed Model for Individual-Organization Alignment
The model proposed in this paper was developed as part of a doctoral research
program [17]. This work makes the case for an enterprise perspective centred on
agents and contexts. More specifically, this research (1) develops an agent-centric
perspective that is complementary to activity, technology, information, and
strategy/organization perspectives, and (2) proposes a way to link the agent
perspective with these perspectives. The concept of context provides the key for this
'linkage'. The proposed view is part of a conceptual framework, integrated by a
layered model of organizational agents, and a methodological approach to build
representations based on this ontology. This framework aims at enriching enterprise
modelling, providing and analytical tool for organizational analysis and (re)design
The model departs from five essential concepts (activity, resource, agent, role, and
context), and integrates agent and enterprise architectures that integrates multiple
concerns of agent behaviour. As a result, agent behaviour is captured in terms of
activity and resource-related roles, which are organized in three layers; (1) action, (2)
deliberation, and (3) change/learn layer. The separation of behavioural concerns in
different layers allows not only the addressing of more complex concerns, but also
defining modes of representation consistent with the complexity level of each
behavioural layer. Moreover, it provides a means of exploring and uncovering the
influence between different concerns.
The methodological approach developed to build representations based on this
model is organized around the notion of context, and encompasses six activities: (1)
bootstrapping, (2) action capture, (3) context discovery, (4) context-based analysis
and (5) context integration .
Fig. 1. The Methodological Approach.
In Figure 1 it can be seen that the bootstrapping phase is used to define the group
of agents to be observed, as well as the basic type of actions and resources to be
captured. The action capture phase collects agent daily actions and the resources
involved in such actions, in their chronological order of execution. Context discovery
aims at identifying and characterizing different personal and inter-personal contexts
from action groupings. Context-based analysis aims at finding recurrent patterns
within personal and inter-personal contexts. Context integrations aims at relating
contextual representations with formal tasks and resources.
Case validation was accomplished using non-structured interviews and
questionnaires, and enabled evaluating not only the benefits, but also the limitations
of the framework.
6.1 Some Proposed Solutions for Organizational Modelling
The ontological position of emergence poses several requirements on organizational
models. The model put forward proposes some solutions that aim at overcoming the
shortcomings of organizational modelling in satisfying such requirements.
Enhanced Traceability of Organizational Agents. Current EM modelling
approaches address organization’s complexity defining several, inter-related
perspectives. Nonetheless, none of these approaches fully acknowledge that agents
themselves are complex entities, which also needs to be handled with their own
architecture. Consequently, EM frameworks provide limited support in addressing
questions about organizational agents. A framework that integrates agent and
organizational architectures and contributes to an uncovering of agent-centric
behaviours is needed.
Situated EM. Current EM approaches assume the existence of unique, external
viewpoints, and produce ’aerial’ representations i.e. representations seen from the
outside of the organization. These representations, while meaningful for some
organizational members, are meaningless or incorrect for others. Enterprise
representations make sense for specific agents, and specific contexts. Departing from
actual actions and interactions, and the inclusion of the notion of context enables
situated enterprise representations. Modelling frameworks that take this fact into
account and allows the modelling of behaviours, situated within specific contexts are
needed. Developing ’context-aware’ enterprise representations provides the
conceptual richness required to address more properly, the complexity of
organizations and their agents. It also enables a proper understanding and comparison
of such representations, as well as their evolution in time.
Model Acquisition from Action Repositories. Enterprise models are mostly built
from interviews, surveys, questionnaires, observation and analysis of textual
descriptions of activities. The requirement is for a model acquisition approach that
allows creating representations from action repositories, an approach that departs
from a discussion of basic action types, and resources, as well as their meanings. This
departure aims at building representation conveying the same message to all
participating agents. In EM achieving consensus around the meaning of activity and
resource names requires is a time consuming processes. The usage of small semantic
units such as actions and resource-related items as the basic building blocks of the
approach eased the process of achieving consensus around their meaning.
Capturing and Modelling Work Practices. Current EM frameworks capture generic
task, activity, and process model that define behaviour at a role level. Modelling work
practices require the capability of answering the question; ”How does Individual i
perform Activity A? Which resource(s) use?”. This compound question has been
addressed by independent research in systems development and simulation, but not by
EM frameworks.
Capturing and modelling work practices means building diagrams situated in
particular contexts, reflecting the particular action types, action flows and resources
employed by given individuals in performing given tasks. Since these resources can
be human, diagrams reflecting inter-personal patterns must be built. This means the
ability to answer questions such as (1) ”Who (Individual i1) interact with who
(Individual i2)?”, and (2) ”How does Individual i1 interact with Individual i2?” These
questions must be addressed using a representation language and model acquisition
approach better fitted for purposes of organization analysis.
Capturing and Modelling Multitasking Behaviour. The impact of human
multitasking in individual productivity has been acknowledged by several researchers.
In these works, multitasking behaviour does not reflect how work is accomplished.
Rather, it reflects how agents manage themselves. It requires the capability to answer
question such as ”How does Individual I manage Resource R?”, where Resource R is
the individual him/herself. This behavioural concern has been addressed in research
works of human-machine interaction, human resource management, cognitive
sciences, but no EM framework has addressed it.
Capturing and modelling multitasking requires using a deliberation layer which
means using the notion of context to define work fragmentation, rather than tasks.
Multitasking behaviour is modelled in terms of context interleaving, and context
activation rules. Different tasks may require similar resources. Likewise, the same
task may require different resources, at different stages. Since switching costs are
caused by the need to ’pull’ different set of physical and cognitive resources, and
contexts reflect resource groupings, this criteria is more appropriate to measure work
fragmentation than tasks.
Aligning Design and Execution. The problem of linking individual behaviours with
organizational activities and resources is disregarded by EM acquisition approaches,
which depart from higher level of abstraction. The problem of aligning organization’s
design with actual execution using action logs, has been acknowledged and addressed
by the process mining research. However, the focus of this field is restricted to the
alignment of pre-defined application workflows, with workflows acquired from
execution data collected from logs produced by WFMS, and enterprise applications.
This work does not collect data from non-structured actions stored in message-based,
groupware applications, where messages are not associated with tasks. It also
disregards non-computer mediated actions and interactions, which require to be
registered manually. Without unstructured, non-classified actions, it is not possible to
get accurate definitions of actual organization workflows.
7 Tooling Algardata for Individual-Organization Alignment
The present case is an on-going case that is serving as a test bed for our approach. The
organizational setting is Algardata, a portuguese IT enterprise created in 1990.
Currently, Algardata employs more than a hundred collaborators, and provides a
variety of IT services. Algardata's clients include banks, government institutions,
hotel chains, distribution enterprises, telecommunication operators, as well as
professional law, architecture and consulting firms. During the last five years,
Algardata has experienced a very fast growth. The high number of knowledge
workers posed two important challenges. The first is related to the definition of
effective productivity measures for these workers. The second is tooling the
organization with appropriate means to identify all the variety of tasks performed,
how and when these tasks are performed, and which human and technological
resources are used. The goal of this work is to implement the proposed approach in
combination with traditional business and task modeling activities. In short, the case
goals were: (1) provide the basis for individual productivity measures, (2) provide the
basis for a bottom-up discovery of individual and group strategies and, (3) uncover
software development workflows.
The first step involved the identification of action logs sources. Three main sources
were identified. The first is the current production control application. Algardata
developed an application the production of software developers, where they introduce
on a daily basis, the time spent on different activities. With minor redesigns, the logs
created by this application are an ideal source of action logs. However, since they are
registered manually, these logs do not capture all worker actions. Hence, additional
sources are required. The second are e-mail logs. Though these logs require additional
and more complex processing in order to identify actions, e-mail logs contain a great
amount of valuable information in uncovering interaction networks. The third source
will be provided the logs from the Microsoft CRM application ®, currently in
The second step involved the redefinition of the log structure of the production
control application. Employees introduce the time spent on project phases. They
introduce the client, project and phase related to that activity. Clients always refer to
clients. Projects belong to a unique client but several projects may be related to a
single client. Each project is divided in several phases. The specific activity
performed is not registered. Though this structure allows extracting some statistics
about the time spent by employees, it has several limitations for the purposes of this
case. The structure assumes a unique hierarchy client -> project -> phase, which does
not offer the flexibility and detail required to analyze the time spent from multiple
perspectives. Activities performed for internal clients are not registered.
The classification of individual actions in the corresponding project phase is
performed by the employees, and does not obey to any scheme or structure
representing the information requirements of the managers. Further, it poses the risk
of inconsistent classifications since different employees may classify differently
similar actions. Finally, the limited analytic possibilities of the actual structure have a
negative impact on employees' motivation to register their actions because they do not
see any benefits in this effort. The logs were thus re-structured to overcome these
problems. First, the new structure acknowledges external and internal clients. Second,
a set of action types will be defined. These types to be defined are semantically closer
to daily actions (e.g. propose, request, promise, elaborate (document)). Action types
are complemented with the resource-related items used or produced by the actions.
These actions will be grouped in contexts. Since activities are abstractions, the
specific relation of contexts and activities will vary according the activity definition.
Contexts are then related with formal activities in a many-to-many fashion, giving the
possibility of relating individual actions to several activities.
This case has recently finished its bootstrapping phase, that is, the definition of the
basic action types, and resource-related items. The identification of basic action and
resource types is based on two main sources; (1) direct observation and (2) analysis of
e-mail logs. An external observer captured the activities of the Aurora-Soft team
during random visits, along a two month-period. A log of six hundred mails was also
manually analyzed in searching for action types employed by the the Aurora-Soft
team. As a result, a set of action was identified that is not included due to space
reasons. A set of context types was also defined, by Algardata's management and will
serve as action grouping criteria.
The description of this case seeks to better illustrate the model proposed.
Nonetheless, this case is in a preliminary phase. The proper methods for context
identification, analysis and integration phases, as well as the tools to support such
methods remain to be discussed, defined and developed. Upon the case completion,
we expect to enhance Algardata’s capability of answering questions about their
workers, particularly of their software developers. Algardata expects to know the
actual action and interaction patterns associated to different developing contexts that
uncover not only how tasks are performed but which specific resources are used or
provided by individuals, and how individuals manage their tasks and resources.
8 Conclusions
Over the years organizational research has identified forces within the organization
which are more enduring and hard to change as opposed to others with are more
ephemeral or amenable to adoption. The former are usually of an informal nature (e.g.
cultural norms) and the later are of a more formal character (e.g. HRIS). From the
point of the view of the researcher these kinds of forces are quite unrelated and often
difficult to reconcile. However, the implementer of systems in the real world knows
that these kinds of forces are related and that neglecting either of them could mean
failure of the project. Hence, when researching HRIS from an organizational and
integrated perspective, as recommended by Ruel and Magalhaes (2008), it is crucial
to combine concepts from the sociology of organizations with techniques from
systems modelling. This is what we have tried to achieve in the present paper.
Social emergence is the ontological point of view which we defend in this paper. In
line with the autopoietic view, Fuchs (2003) argues that society can only be explained
consistently as self-reproducing if man is recognised as a social being and has a central
role in the reproduction process. Through social actions, social structures are constituted
and differentiated, meaning that social interaction makes new qualities and structures
emerge which cannot be reduced to the individual level. This is a process of bottom-up
emergence which we believe lies also at the root of the process of organizational
engineering. If we concur that the study of HRIS has crucial organizational
implications, then the study of HRIS should encompass the engineering and modelling
considerations we have put forward in this paper.
We have put a conceptual framework whose main purpose is to facilitate the
alignment between individuals and organizations. Nonetheless, organizations have
several levels of complexity, which are typically structured around individual, inter-
personal, group, organization-wide levels, as well as inter-organizational levels.
Hence, aligning individuals and organizations need to be accomplished on a level-by-
level basis. The proposed framework defines an approach to align individual and
inter-personal views with group-level views. At these levels, alignment entails
addressing several concerns of individual and inter-personal behaviour, and relating
these behaviours with individual tasks, organizational activities and resources.
We have illustrated the framework by means of a case study which aims at
highlighting some crucial modelling criteria of emergent phenomena such as
organizations or HRIS in organizations. The key modelling propositions contained in
this framework are as follows: (1) Enhanced traceability of organizational agents, (2)
Situated enterprise modelling, (3) Model acquisition from action repositories, (4)
Capturing and modelling work practices, (5) Aligning design and execution.
1. Ruel, H.; Magalhaes, R. : Human Resource Information Systems: an Integrated Research
Agenda. Forthcoming book chapter, Elsevier Science (2008).
2. Sawyer, R. K.: Social Emergence: societies as complex systems. Cambridge, UK:
Cambridge University Press (2005).
3. Mingers, J. “Information, Meaning and Communication: an autopoietic approach” in F.
Geyer; J. van der Zouwen (ed) Sociocybernetics: complexity, autopoiesis and observation
of social systems.Westport, CT: Greenwood Press (2001).
4. Boulding, K.E.: “General Systems Theory: the skeleton of science.” Management Science,
2: 197-208. (1956).
5. Varela, F.J.: “Two Principles of Self-Organization,” in H. Ulrich; G.J.B. Probst (ed.), Self
Organization and Management of Social Systems. N. York: Springer Verlag. (1984).
6. Hill, W. F.: Learning: a survey of psychological interpretations. N. York: Longman (1997).
7. Varela, F.J.; Thompson, E.; Rosch, E.: The Embodied Mind. Cambridge, MA: MIT Press
8. Varela, F.J.: “Whence Perceptual Meaning? A Cartography of Current Ideas,” in F.J.
Varela; J-P Dupuy (ed.), Understanding Origins. Dordrecht, Holland: Kluwer Academic
Publishers (1992).
9. Giddens, A.: The Constitution of Society: outline of the theory of structuration. Cambridge,
UK: Polity Press (1984).
10. Di Paolo, E.: “Autopoiesis, Adaptivity, Teleology, Agency”. Phenomenology and the
Cognitive Sciences, 4: 429-452 (2005).
11. De Jaegher, H.; Di Paolo, E.: “Participatory Sensemaking: an enactive approach to social
cognition”. Phenomenology and the Cognitive Sciences, 6: 485-507 (2007).
12. Dignum, V.: A model for organizational interaction: based on agents, founded on logic.
SIKS Dissertation Series Nº 2004-1. PhD Thesis, Utrecht University, Netherlands (2004)
13. Liles, D.H.; Presley, A.: Enterprise modelling within an enterprise engineering framework.
Winter Simulation Conference, Association for Computing Machinery), San Diego, CA,
USA (1996).
14. Hatch, M. J.; Cunliffe, A. L. Organization Theory. Oxford: Oxford University Press (2006)
15. Caetano, A.; Silva, A.; Tribolet, J. “Separation of concerns in object-oriented business
process modeling”. 7th International Conference on Information Systems Implementation
Modelling, Rožnov pod Radhošťem, Czech Republic (2004).
16. Sakkerman, J.: How to Survive the Jungle of Enterprise Architecture. Trafford (2004)
17. Zacarias, M. S.: A Conceptual Framework for the Alignment of Individuals and
Organizations. Unpublished PhD Dissertation (submitted). Lisbon, Portugal: Instituto
Superior Tecnico (2008).
18. Fuchs, C.: “Structuration Theory and Self-Organization”. Systemic Practice and Action
Research, 16 (2), 133-167 (2003).