Moving Across Paradigms between the Process Design and Enactment

Phase in Enterprise Information Systems

Filip Caron and Jan Vanthienen

Leuven Institute for Research on Information Systems (LIRIS), Faculty of Business and Economics, KU Leuven,

Naamsestraat 69, 3000 Leuven, Belgium

Keywords:

Process-aware Information Systems, Business Process Management, Process Modeling, Process Enactment,

Transitions, Enterprise Information Systems.

Abstract:

While the business process management literature often assumes a single approach (e.g. procedural or event-

driven) over the process lifecycle, a transition between approaches at different phases in the process lifecycle

may signiﬁcantly reduce the impact of intrinsic trade-offs between process characteristics. This position paper

explores several business process strategies by analyzing the approaches at different phases in the process

lifecycle as well as the various transitions.

1 INTRODUCTION

Organizations face a continuous pressure to improve

process compliance, ﬂexibility, efﬁciency and effec-

tiveness. While responding to these pressures indi-

vidually can be demanding, the real challenge is deal-

ing with the intrinsic tradeoffs (e.g. between com-

pliance, efﬁciency and ﬂexibility). A wide spectrum

of business process management paradigms has been

presented in the literature, each with its speciﬁc trade-

offs. Additionally, most of these contributions con-

sider the selection of the optimal approach given the

business environment as a one-time choice at the pro-

cess design phase. However, business processes may

also require different tradeoffs at different phases of

the process lifecycle.

The contribution of this position paper will be the

exploration of various business process strategies that

combine the selection of a design-time paradigm and

a run-time paradigm (i.e. position selection) with

a transition path, resulting in a better ﬁt between

the business processes and the business environment.

Rather than making a value judgement of the different

strategies, we will focus on a discussion of the impact

on the process characteristics.

The remainder of this paper is structured as fol-

lows: ﬁrst we introduce and assess the different po-

sitions (section 2 & 3), followed by the speciﬁcation

and analysis of the transition strategies (section 4) and

section 5 concludes the paper.

2 THE BUSINESS PROCESS

PARADIGMS AND LIFECYCLE

Each position in the process lifecycle is deﬁned by the

combination of the related lifecycle phase and process

paradigm.

Traditionally, the traditional business process life-

cycle consists of four phases with distinct roles

(Weske, 2007), i.e. process design phase, process

implementation phase, process enactment phase and

process evaluation phase. The design phase and the

enactment phase straightforwardly belong to respec-

tively the design-time and run-time. As the process

analysis and evaluation phase often results in spe-

ciﬁc recommendations/requirements for a process re-

design, it can be considered as part of the next cycle’s

design-time. The implementation phase can be asso-

ciated with run-time as it is the implemented process

that will be executed.

The business process management literature has

proposed a wide spectrum of process paradigms with

at the extremes the procedural (Zisman, 1977; OMG,

2006; Ellis and Nutt, 1993) and declarative (Goed-

ertier and Vanthienen, 2009; van der Aalst et al.,

2009b; Swenson, 2010) process paradigm. In be-

tween there exists a wide variety of hybrid paradigms

that combine aspects of both extremes (Sadiq et al.,

2005; van der Aalst et al., 2009a; Schonenberg et al.,

2008; Kumar and Yao, 2009; Hallerbach et al., 2010).

218

Caron F. and Vanthienen J..

Moving Across Paradigms between the Process Design and Enactment Phase in Enterprise Information Systems.

DOI: 10.5220/0003999202180223

In Proceedings of the 14th International Conference on Enterprise Information Systems (ICEIS-2012), pages 218-223

ISBN: 978-989-8565-12-9

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

3 ASSESSING DESIGN-TIME

AND RUN-TIME POSITIONS

Since every position has its own design or execu-

tion principles, all positions have different character-

istics and consequently appeal to different business

requirements. In (Davenport, 1993) Davenport iden-

tiﬁes four desirable qualities for business processes;

process ﬂexibility, compliance, effectiveness and efﬁ-

ciency. This section analyzes the possible impact of

each position’s principles on these characteristics.

• Process ﬂexibility is the extent to which an or-

ganization can deal with business process change,

the ability to accommodate the special needs of

particular business process instances (i.e. run-

time ﬂexibility) as well as to accommodate pro-

cess model evolutions (i.e. maintainability).

• Process compliance is the extent to which a pro-

cess is in correspondence with the internally de-

ﬁned business rules and the externally imposed

business regulations.

• Process effectiveness is the extent to which a

business process realizes its business goals.

• Process efﬁciency is the extent to which the or-

ganization of the business process is capable of

minimizing the amount of utilized resources such

as personnel, materials, time, machine capacity.

Additionally, two important characteristics of pro-

cess modeling languages are taken into account for

the evaluation of the design-time positions:

• The expressibility of a process modeling lan-

guage is determined by its ability to express spe-

ciﬁc process elements, e.g. control-ﬂow, data, ex-

ecution and temporal information (Lu and Sadiq,

2007; zur Muehlen et al., 2007)

• The level of comprehensibility reﬂects the ability

of a process modeling language to deﬁne under-

standable process models that can be easily com-

municated among various stakeholders (Fahland

et al., 2009a).

Tables 1 and 2 represent the assessment results of

respectively the design-time and run-time positions.

4 DESIGN-TIME TO RUN-TIME

TRANSITIONS: DEFINITION

AND ASSESSMENT

Traditional business process management solutions

are oriented towards a single process paradigm, e.g.

the business processes are modeled using procedural

process modeling languages (such as BPMN (OMG,

2006)) and then executed in a procedural enactment

environment (such as BPEL (OASIS, 2007)). These

design-time to run-time transitions within a single

process paradigm are the same paradigm transitions.

There also exist business processes that have differ-

ent requirements at design-time and run-time, e.g.

in terms of ﬂexibility. In these cases we analyze

design-time to run-time transitions between process

paradigms, the cross paradigm transitions.

Different process lifecycle phases may require dif-

ferent tradeoffs between process characteristics. Con-

sequently, cross paradigm transitions may in practice

result in a better ﬁt between the business process-

aware information systems and the business environ-

ment.

4.1 Same Paradigm Transitions

Procedural - Procedural Transition. Different

transformation strategies between procedural process

modeling languages and procedural process execu-

tion languages have been proposed in the literature

(Decker et al., 2008; Ouyang et al., 2009). How-

ever, due to a conceptual mismatch between the stan-

dard procedural process modeling and execution lan-

guages, translation techniques are generally only of-

fered for a core subset of the procedural process mod-

eling constructs (Recker and Mendling, 2006).

Impact on Process Characteristics. A set of research

contributions on process variant management (based

on querying process variants before run-time) falls

into the scope of the procedural-procedural transi-

tion, e.g. (Lu et al., 2009). While this procedural-

procedural transition might have a positive impact on

the process ﬂexibility, managing changing compli-

ance requirements can be challenging for repositories

with large collections of process variants. The im-

pact on process efﬁciency and effectiveness depends

on the quality of the process variants and the number

of different situations captured by the process variants

in the repository. In (Reichert et al., 2009) a technique

is presented for adding, replacing and moving activi-

ties.

Example Process Case. The procedural-procedural

transition can be recommendedfor business processes

in a stable environment with predictable execution

paths, e.g. processing standard and static items (such

as online orders).

Declarative - Declarative Transition. The common

declarative process modeling languages have their

roots in formal logic. Due to this formal founda-

tion the translation of a high-level declarative process

MovingAcrossParadigmsbetweentheProcessDesignandEnactmentPhaseinEnterpriseInformationSystems

219

Table 1: Design-Time Positions Analysis.

Procedural process

modeling

Declarative process

modeling

Hybrid process modeling

Description Exact speciﬁcation of all ex-

ecution paths, events and

exceptions.

Set of constraints, business

rules, event conditions or

other (logical) expressions

that deﬁne mandatory prop-

erties and dependencies be-

tween activities.

Combination of procedural

and declarative constructs

resulting in process models

with placehoder activities or

procedural process models

with rule-based adaption

Impact on process

characteristics

Positive: Comprehensible

models, process efﬁciency

Negative: Overspeciﬁcation

(Sadiq et al., 2005), main-

tainability issues (Fahland

et al., 2009b)

Positive: Flexibility,

compliance by design,

maintainability (traceability

and absence of duplication),

high expressibility (Lu and

Sadiq, 2007)

Negative: Limited com-

prehensibility (large and

unstructured sets of formal

rules) (Fickas, 1989)

Moderation of impact:

Improved ﬂexibility and

reduced maintainability

issues (declarative place-

holder activities), Improved

efﬁciency and reduced

ﬂexibility (procedural

placehoder activities),

improved ﬂexibility (rule-

based adaption)

Languages examples Petri Net (based) modeling

(Zisman, 1977; Ellis and

Nutt, 1993), BPMN (OMG,

2006) and UML Activity

Diagram (OMG, 2004)

ConDec (Pesic and van der

Aalst, 2006), DecSerFlow

(van der Aalst and Pesic,

2006) and BPCN (Lu et al.,

2009)

Placeholder activities

(Sadiq et al., 2005; van der

Aalst et al., 2009a; Scho-

nenberg et al., 2008),

rule-based adaption (Kumar

and Yao, 2009; Hallerbach

et al., 2010)

Table 2: Run-Time Positions Analysis.

Procedural process

enactment

Declarative process

enactment

Hybrid process enactment

Description Straightforward execution

based on precisely speciﬁed

execution paths

An individual execution

path that satisﬁes the set

of mandatory constraints is

dynamically created

The base process is ex-

ecuted according to the

principles of the paradigm,

whenever a placeholder

activity is executed a

paradigms switch takes

place

Impact on process

characteristics

Positive: high efﬁciency

Negative: limited ﬂexibility

Positive: high ﬂexibility, as-

sured compliance

Negative: little support, lim-

ited efﬁciency

Characteristics of process

parts are determined by

their paradigm.

Flexibility as a service im-

proves ﬂexibility

Language examples & ex-

ecution environments

BPEL (OASIS, 2007) and

YAWL (van der Aalst and

Ter Hofstede, 2005)

LTL (Pesic et al., 2008),

CTL (Yu et al., 2006),

PLMﬂow (Zeng et al.,

2002), ECA rules (Kappel

et al., 1998) and event rules

(Paschke and Boley, 2009)

Chameleon (Sadiq et al.,

2005) and subprocesses

encapsulated in a service

(van der Aalst et al., 2009a)

model into enactable rules is rather straightforward

(e.g. translation of a ConDec process model into LTL

expressions (Pesic et al., 2008)).

Impact on Process Characteristics. During the

declarative-declarative transition no (additional) fac-

tors that affect the business process’s ﬂexibility, com-

pliance, efﬁciency or effectiveness will/can be intro-

duced.

Example Process Case. These transitions are suitable

for business processes in a highly evolving environ-

ment and/or business processes with non-predictable

execution paths. As declarative process management

systems might provide limited support at run-time

(Weber et al., 2009), this transition type will be most

suited for experts dealing with unique cases (Schmidt,

2006) (e.g. non-standardized health care processes).

Hybrid - Hybrid Transition. The hybrid-hybrid

transition can only be applied on hybrid process mod-

els that contain placeholder activities. Within the con-

text of this type of hybrid paradigm, the business

process paradigms of each process parts determine

which same paradigm transition will be used for that

ICEIS2012-14thInternationalConferenceonEnterpriseInformationSystems

220

process part.

Impact on Process Characteristics. As these tran-

sitions are intrinsically similar to the previously de-

scribed same strategy transitions, we argue that the

impact on the process characteristics of this transi-

tion is determined by those same strategy transitions.

It should be noted that process variant management

approaches in this context will be easier to main-

tain than those presented in the context of procedural-

procedural transitions, since the base structure is not

duplicated.

Example Process Case. Hybrid-hybrid transitions

will be used for business processes that contain both

process parts with stable and highly evolving environ-

ments and/or that consist of both process parts with

predictable and non-predictable execution paths (e.g.

an advisory project with structured administrative ac-

tivities and unstructured problem solving parts).

4.2 Cross-Paradigm Transitions

While the impact of the same paradigm transitions on

the desirable characteristics of a business process is

rather limited, the impact of the cross-paradigm tran-

sitions can be rather extensive.

Procedural - Declarative Transition. The procedu-

ral process model is translated into a set of event-

based business rules (e.g. preconditions), which can

be used for a declarative process enactment (Casati

et al., 1998).

Impact on Process Characteristics. Since the im-

plicit constraints governing the procedural process

model are exactly mapped on event-based business

rules, the issue of overspeciﬁcation is not dealt with.

However, process ﬂexibility slightly increases com-

pared to the procedural-procedural transition due to

the possibility of a run-time replacement or addition

of a service task and the ability to deﬁne extra event-

based business rules to deal with temporary circum-

stances (Dumas et al., 2005). The process effective-

ness and efﬁciency are determined by the quality of

the process model and the declarative execution envi-

ronments.

Example Process Case. The procedural-declarative

transition is suitable for use within the context of dis-

tributed processes, for which the process environment

remains relatively stable and the ability to dynam-

ically deal with temporary circumstances is valued

(e.g. order-to-cash process in virtual organizations).

Declarative - Procedural Transition. Before run-

time a systematic procedure is used for the construc-

tion of an optimal control-ﬂow with reference to a

particular characteristic. This procedure is closely

related to artiﬁcial intelligence planning techniques

(Hendler et al., 1990; Ferreira and Ferreira, 2005).

Impact on Process Characteristics. While the

declarative process speciﬁcations provide extensive

design-time ﬂexibility, run-time ﬂexibility remains

limited to the ﬂexibility offered by procedural en-

actment. However, the declarative process model in

combination with a time-efﬁcient planning algorithm,

allows for a rapid adoption of new compliance re-

quirements. Moreover,when the proceduralworkﬂow

engine does not support any of the run-time ﬂexibil-

ity enhancing techniques, compliance can be easily

checked against the declarative process model.

The use of an artiﬁcial planning algorithm might

positively affect both the process efﬁciency and ef-

fectiveness, since an optimization criterion needs to

be speciﬁed. In addition, compared to declarative

process enactment the end-user will be sufﬁciently

guided and supported.

Example Process Case. The declarative-procedural

transition is useful for processes that require far-

reaching redesigns at regular intervals and/or for pro-

cesses that beneﬁt from an optimization with refer-

ence to a certain criterium. These process, however,

are at the same time relatively stable in the periods

between those redesign phases.

Hybrid - Procedural Transition. Within the con-

text of this transition the focus is primarily placed on

hybrid models of the second type, the process mod-

els that combine a full procedural speciﬁcation with

a set of business rules. Before run-time the procedu-

ral reference model is customized to the speciﬁc needs

of a particular case by applying the set of customiz-

ing business rules (Kumar and Yao, 2009; Hallerbach

et al., 2010).

Impact on Process Characteristics. Due to the hy-

brid process model as a starting position, a neat ap-

proach to process variant management is provided.

Compared to the process variant management ap-

proach introduced in the procedural-procedural tran-

sition, maintenance of requirements is not needlessly

complicated since there is no duplication of the base

process. However, the customization must be per-

formed correctly and completely in order not to affect

the process effectiveness and compliance.

Example Process Case. This type of hybrid - proce-

dural transition will most likely be used for support-

ing a set of business processes that all only slightly

differ from a speciﬁc reference process. We also ex-

pect the business processes to have predictable execu-

tion paths (e.g. claim handling for different insurance

MovingAcrossParadigmsbetweentheProcessDesignandEnactmentPhaseinEnterpriseInformationSystems

221

products).

Translating Declarative Placeholders. Naturally, hy-

brid process models with placeholder activities can be

transformed into procedural execution models as de-

scribed in the declarative-proceduraltransition, which

also results in a hybrid-procedural transition. Charac-

teristics of the transition in this context are compara-

ble with those of the declarative-proceduraltransition.

5 CONCLUSIONS

Designing information systems that provide support

for operational business processes with the right level

of process ﬂexibility, compliance, efﬁciency and ef-

fectiveness can be a challenging task. This posi-

tion paper promotes a clear distinction between the

business process strategies and their differences at

distinct points in the process life cycle. Further-

more, the paper elaborated on the transitions between

design-time and run-time; in addition to the same

paradigm transitions three interesting cross-paradigm

transitions were presented.

REFERENCES

Casati, F., Ceri, S., Pernici, B., and Pozzi, G. (1998).

Deriving active rules for workﬂow enactment. In

Database and Expert Systems Applications, pages 94–

115. Springer.

Davenport, T. (1993). Process innovation: reengineering

work through information technology. Harvard Busi-

ness Press.

Decker, G., Dijkman, R., Dumas, M., and Garc´ıa-Ba˜nuelos,

L. (2008). Transforming BPMN diagrams into YAWL

nets. Business Process Management, pages 386–389.

Dumas, M., Fjellheim, T., Milliner, S., and Vayssi`ere,

J. (2005). Event-Based Coordination of Process-

Oriented Composite Applications. Business Process

Management, pages 236–251.

Ellis, C. and Nutt, G. (1993). Modeling and enactment of

workﬂow systems. Application and Theory of Petri

Nets 1993, pages 1–16.

Fahland, D., Lubke, D., Mendling, J., Reijers, H., Weber,

B., Weidlich, M., and Zugal, S. (2009a). Declar-

ative versus imperative process modeling languages:

The issue of understandability. Enterprise, Business-

Process and Information Systems Modeling, pages

353–366.

Fahland, D., Mendling, J., Reijers, H., Weber, B., Weidlich,

M., and Zugal, S. (2009b). Declarative vs. Imperative

Process Modeling Languages: The Issue of Maintain-

ability. In 1st International Workshop on Empirical

Research in Business Process Management, pages 65–

76, Ulm, Germany.

Ferreira, D. and Ferreira, H. (2005). Learning, planning,

and the life cycle of workﬂow management. In EDOC

Enterprise Computing Conference, 2005 Ninth IEEE

International, pages 39–45. IEEE.

Fickas, S. (1989). Design issues in a rule-based system.

Journal of Systems and Software, 10(2):113–123.

Goedertier, S. and Vanthienen, J. (2009). An overview

of declarative process modeling principles and lan-

guages, volume 6, pages 51–58. Communications of

systemics and informatics world network.

Hallerbach, A., Bauer, T., and Reichert, M. (2010). Captur-

ing variability in business process models: the provop

approach. Journal of Software Maintenance and Evo-

lution: Research and Practice, 22(6-7):519–546.

Hendler, J., Tate, A., and Drummond, M. (1990). AI plan-

ning: Systems and techniques. AI magazine, 11(2):61.

Kappel, G., Rausch-Schott, S., and Retschitzegger, W.

(1998). Coordination in workﬂow management sys-

tems a rule-based approach. Coordination Technology

for Collaborative Applications, pages 99–119.

Kumar, A. and Yao, W. (2009). Process Materialization

Using Templates and Rules to Design Flexible Pro-

cess Models. In Proceedings of the 2009 Interna-

tional Symposium on Rule Interchange and Applica-

tions, pages 122–136. Springer-Verlag.

Lu, R. and Sadiq, S. (2007). A survey of comparative busi-

ness process modeling approaches. In Business Infor-

mation Systems, pages 82–94. Springer.

Lu, R., Sadiq, S., and Governatori, G. (2009). On manag-

ing business processes variants. Data & Knowledge

Engineering, 68(7):642–664.

OASIS (2007). Web services business process execution

language 2.0.

OMG (2004). Uml 2.0 superstructure speciﬁcation.

OMG (2006). Business process modeling notation (bpmn)

1.1.

Ouyang, C., Dumas, M., Aalst, W., Hofstede, A., and

Mendling, J. (2009). From business process models to

process-oriented software systems. ACM transactions

on software engineering and methodology (TOSEM),

19(1):1–37.

Paschke, A. and Boley, H. (2009). Handbook of Research

on Emerging Rule-Based Languages and Technolo-

gies: Open Solutions and Approaches, chapter Rules

capturing events and reactivity. IGI Publishing.

Pesic, M., Aalst, W., and Eijnatten, F. (2008). Constraint-

based workﬂow management systems: shifting control

to users. PhD thesis, Technische Universiteit Eind-

hoven.

Pesic, M. and van der Aalst, W. (2006). A declarative ap-

proach for ﬂexible business processes management.

In Business Process Management Workshops, pages

169–180. Springer.

Recker, J. and Mendling, J. (2006). On the translation be-

tween bpmn and bpel: Conceptual mismatch between

process modeling languages. In The 18th Interna-

tional Conference on Advanced Information Systems

Engineering. Proceedings of Workshops and Doctoral

Consortium, pages 521–532.

ICEIS2012-14thInternationalConferenceonEnterpriseInformationSystems

222

Reichert, M., Rinderle-Ma, S., and Dadam, P. (2009). Flexi-

bility in process-aware information systems. Transac-

tions on Petri Nets and Other Models of Concurrency

II, pages 115–135.

Sadiq, S., Orlowska, M., and Sadiq, W. (2005). Speciﬁca-

tion and validation of process constraints for ﬂexible

workﬂows. Information Systems, 30(5):349–378.

Schmidt, R. (2006). Flexibility in service processes. In

Proceedings of the CAISE 2006 Workshop on Busi-

ness Process Modelling, Development and Support,

BPMDS.

Schonenberg, H., Mans, R., Russell, N., Mulyar, N., and

Aalst, W. (2008). Process ﬂexibility: A survey of con-

temporary approaches. Advances in Enterprise Engi-

neering I, pages 16–30.

Swenson, K. (2010). Mastering the unpredictable: How

adaptive case management will revolutionize the way

that knowledge workers get things do.

van der Aalst, W., Adams, M., Hofstede, A., Pesic, M., and

Schonenberg, H. (2009a). Flexibility as a Service. In

Database Systems for Advanced Applications, pages

319–333. Springer-Verlag.

van der Aalst, W. and Pesic, M. (2006). DecSerFlow: To-

wards a truly declarative service ﬂow language. Web

Services and Formal Methods, pages 1–23.

van der Aalst, W., Pesic, M., and Schonenberg, H. (2009b).

Declarative workﬂows: Balancing between ﬂexibility

and support. Computer Science-Research and Devel-

opment, 23(2):99–113.

van der Aalst, W. and Ter Hofstede, A. (2005). YAWL:

yet another workﬂow language. Information Systems,

30(4):245–275.

Weber, B., Sadiq, S., and Reichert, M. (2009). Beyond

rigidity–dynamic process lifecycle support. Computer

Science-Research and Development, 23(2):47–65.

Weske, M. (2007). Business Process Management: Con-

cepts, Languages, Architectures. Springer-Verlag

New York Inc.

Yu, J., Manh, T., Han, J., Jin, Y., Han, Y., and Wang, J.

(2006). Pattern based property speciﬁcation and ver-

iﬁcation for service composition. Web Information

Systems–WISE 2006, pages 156–168.

Zeng, L., Flaxer, D., Chang, H., and Jeng, J. (2002).

PLM ﬂowDynamic Business Process Composition

and Execution by Rule Inference. Technologies for

E-Services, pages 51–95.

Zisman, M. (1977). Representation, speciﬁcation and au-

tomation of ofﬁce procedures. PhD thesis, Wharton

School.

zur Muehlen, M., Indulska, M., and Kamp, G. (2007). Busi-

ness process and business rule modeling: A repre-

sentational analysis. In EDOC Conference Workshop,

2007. EDOC’07. Eleventh International IEEE, pages

189–196. IEEE.

MovingAcrossParadigmsbetweentheProcessDesignandEnactmentPhaseinEnterpriseInformationSystems

223