PurposeIncreasing flexibility and complexity of modern process‐aware information systems (PAIS) usually leads to less guidance for its users and consequently requires more experienced users. A flexible PAIS allows users to freely choose a specific execution sequence of activities. However, there are no guarantees that the chosen sequence conforms to established business rules. This context proposes a supervisory control service (SCS), which can be used to support end‐users of flexible PAIS during process execution by giving a list of disabled (or enabled) events (activities), i.e. at any point in time a list of possible next steps is given.Design/methodology/approachThe SCS has been implemented in process mining framework (ProM – www.processmining.org) and the experiment that has been performed shows its feasibility. The computational infrastructure already implemented in ProM allows communication with external applications and allows the SCS to be integrated with PAIS that records events. This paper demonstrated that the SCS in ProM can cooperate with PAIS system in order to support users.FindingsThe SCS proposed in this paper monitors and restricts execution sequences of activities such that business rules are always obeyed. The system offers support based on business rules but does not limit the user by imposing rigid control‐flow structures. Users can adopt this service as a guide to execute activities with a guarantee that business rules are followed and goals are met. The SCS also offers flexibility to users to choose execution sequences.Originality/valueThe paper argues that the proposed approach has many advantages on controlling/supporting flexible processes. Supervisory control theory allows an automatic control synthesis instead of the usual manual and heuristic procedures. Thus, new control actions may be rapidly and automatically designed when modifications, such as redefinition of business rules or activities arrangements, are necessary. The flexible processes can be made to behave optimally with respect to a variety of criteria, where optimal means in minimally restrictive way.

Service-oriented computing, an emerging paradigm for architecting and implementing business collaborations within and across organizational boundaries, is currently of interest to both software vendors and scientists. While the technologies for implementing and interconnecting basic services are reaching a good level of maturity, modeling service interaction from a global viewpoint, that is, representing service choreographies, is still an open challenge. The main problem is that, although declarativeness has been identified as a key feature, several proposed approaches specify choreographies by focusing on procedural aspects, leading to over-constrained and over-specified models. To overcome these limits, we propose to adopt DecSerFlow, a truly declarative language, to model choreographies. Thanks to its declarative nature, DecSerFlow semantics can be given in terms of logic-based languages. In particular, we present how DecSerFlow can be mapped onto Linear Temporal Logic and onto Abductive Logic Programming . We show how the mappings onto both formalisms can be concretely exploited to address the enactment of DecSerFlow models, to enrich its expressiveness and to perform a variety of different verification tasks. We illustrate the advantages of using a declarative language in conjunction with logic-based semantics by applying our approach to a running example.

In our earlier work we proposed using the declarative language DecSerFlow for modeling, analysis and enactment of processes in autonomous web services. DecSerFlow uses constraints specified with Linear Temporal Logic (LTL) to implicitly define possible executions of a model: any execution that satisfies all constraints is possible. Hence, a finite representation of all possible executions is retrieved as an automaton generated from LTL-based constraints. Standard model-checking algorithms for creating Büchi automata from LTL formulas are not applicable because of the requirements posed by the proper execution of DecSerFlow (and LTL-based process engines). On the one hand, LTL handles infinite words where each element of the word can refer to zero or more propositions. On the other hand, each execution of a DecSerFlow model is a finite sequence of single events. In this paper we adopt an existing approach to finite-word semantics of LTL and propose the modifications of LTL and automata generation algorithm needed to handle occurrences of single events. Besides eliminating errors caused by the 'multiple properties - single events' mismatch, the proposed adjustments also improve the performance of the automata generation algorithms dramatically.

Traditionally, process mining has been used to extract models from event logs and to check or extend existing models. This has shown to be useful for improving processes and their IT support. Process mining techniques analyze historic information hidden in event logs to provide surprising insights for managers, system developers, auditors, and end users. However, thus far, process mining is mainly used in an offline fashion and not for operational decision support. While existing process mining techniques focus on the process as a whole, this paper focuses on individual process instances (cases) that have not yet completed. For these running cases, process mining can used to check conformance, predict the future, and recommend appropriate actions. This paper presents a framework for operational support using process mining and details a coherent set of approaches that focuses on time information. Time-based operational support can be used to detect deadline violations, predict the remaining processing time, and recommend activities that minimize flow times. All of this has been implemented in ProM and initial experiences using this toolset are reported in this paper.

In order to be competitive, companies are demanding more flexibility from their Process Aware Information Systems (PAISs). However, increasing flexibility and complexity of modern PAISs usually leads to less guidance for its users and consequently requires more experienced users. For instance, a flexible PAIS allows users to freely choose a specific execution sequence. However, there are no guarantees that the chosen sequence is the best one or at least it conforms to established business rules. In this paper we propose a supervisory control service architecture, which can be used to support end users of flexible PAISs during process execution by giving a list of disabled (or enabled) events (activities) i.e. at any point in time a list of possible next steps is given. To evaluate the correctness of the supervisory control service the approach has been implemented and tested.

The lack of flexibility is often seen as an inhibitor for the successful application of workflow technology. Many researchers have proposed different ways of addressing this problem and some of these ideas have been implemented in commercial systems. However, a "one size fits all" approach is likely to fail because, depending on the situation (i.e., characteristics of processes and people involved), different types of flexibility are needed. In fact within a single process/organisation varying degrees of flexibility may be required, e.g., the front-office part of the process may require more flexibility while the back-office part requires more control. This triggers the question whether different styles of flexibility can be mixed and integrated into one system. This paper proposes the Flexibility as a Service (FAAS) approach which is inspired by the Service Oriented Architecture (SOA) and our taxonomy of flexibility. Activities in the process are linked to services. Different services may implement the corresponding activities using different workflow languages. This way different styles of modelling may be mixed and nested in any way appropriate. This paper demonstrates the FAAS approach using the YAWL, DECLARE, and WORKLET services.

Mainstream workflow management systems are using procedural languages ranging from BPMN and EPCs to BPEL and YAWL. By demonstrating DECLARE, we will show that it is also possible to use a fundamentally different approach based on constraints. DECLARE allows for multiple constraint-based languages whose semantics are grounded in Linear Temporal Logic (LTL). The DECLARE system provides a broad range of functionalities ranging from design, enactment and dynamic change to verification, discovery and recommendation. This demo presents the main functionalities of DECLARE and is intended for both researchers and practitioners interested in innovative BPM solutions for processes that require flexibility.

Today’s process-aware information systems tend to either support business processes or provide flexibility. Classical workflow management systems offer good process support as long as the processes are structured and do not require much flexibility. Information systems that allow for flexibility have a tendency to lack process-related support. If systems offer guidance, then they are typically also inclined to ‘‘enforce guidelines’’ and are perceived as inflexible. Moreover, implementing flexible systems is far from trivial. This paper will show that using a more declarative approach can assist in a better balance between flexibility and support. This is demonstrated by presenting the Declare framework that aims to take care of the full spectrum of flexibility while at the same time supports the user using recommendations and other process-mining-based diagnostics.

Recent analysis of clinical Computer-Interpretable Guideline (CIG) modelling languages from the perspective of the control-flow patterns has revealed limited capabilities of these languages to provide flexibility for encoding and executing clinical guidelines. The concept of flexibility is of major importance in the medical-care domain since no guarantee can be given on predicting the state of patients at the point of care. In this paper, we illustrate how the flexibility of CIG modelling languages can be improved by describing clinical guidelines using a declarative approach. We propose a CIGDec language for modelling and enacting clinical guidelines.

The degree of flexibility of workflow management systems heavily influences the way business processes are executed. Constraint-based models are considered to be more flexible than traditional models because of their semantics: everything that does not violate constraints is allowed. Although constraint-based models are flexible, changes to process definitions might be needed to comply with evolving business domains and exceptional situations. Flexibility can be increased by run-time support for dynamic changes – transferring instances to a new model – and ad-hoc changes – changing the process definition for one instance. In this paper we propose a general framework for a constraint-based process modeling language and its implementation. Our approach supports both ad-hoc and dynamic change, and the transfer of instances can be done easier than in traditional approaches.

BPEL has emerged as the de-facto standard for implement- ing processes based on web services while formal languages like Petri nets have been proposed as an "academic response" allowing for all kinds of analysis. Although languages such as BPEL and Petri nets can be used to describe service flows, they both tend to "overspecify" the process and this does not fit well with the autonomous nature of services. Therefore, we propose DecSerFlow as a Declarative Service Flow Language .B y us- ing a more declarative style, there is no need to overspecify service flows. The declarative style also makes DecSerFlow an ideal language for mon- itoring web services, i.e., using process mining techniques it is possible to check the conformance of service flows by comparing the DecSerFlow specification with reality. This can be used to expose services that do not follow the rules of the game. This is highly relevant given the autonomous nature of services.

Workflow management systems support business processes and are driven by their models. These models cover different perspectives including the control-flow, resource, and data perspectives. This paper focuses on the resource perspective, i.e., the way the system distributes work based on the structure of the organization and capabilities/qualifications of people. Contemporary workflow management systems offer a wide variety of mechanisms to support the resource perspective. Because the resource perspective is essential for the applicability of such systems, it is important to better understand the mechanisms and their interactions. Our goal is not to evaluate and compare what different systems do, but to understand how they do it. We use Colored Petri Nets (CPNs) to model work distribution mechanisms. First, we provide a basic model that can be seen as a reference model of existing workflow management systems. This model is then extended for three specific systems (Staffware, FileNet, and FLOWer). Moreover, we show how more advanced work distribution mechanisms, referred to as resource patterns, can be modelled and analyzed.

Recent analysis of clinical Computer-Interpretable Guideline (CIG) modelling languages from the perspective of the control-flow patterns has revealed limited capabilities of these languages to provide flexibility for encoding and executing clinical guidelines [15]. The concept of flexibility is of major importance in the medical-care domain since no guarantee can be given on predicting the state of patients at the point of care. In this paper, we illustrate how the flexibility of CIG modelling languages can be improved by describing clinical guidelines using a declarative approach. We propose a CIGDec language for modelling and enacting clinical guidelines.

In this presentation we shall review different ways of describing the processes of delivering patient care, and relate these to traditional workflow in business processes, and the concept of “careflow” proposed by Panzarasa and her colleagues in Pavia. We shall discuss the problem of designing a careflow application as a form of process modelling, to be contrasted with older paradigms ranging from rule-based alerts and reminders to Petri nets and Critical path analysis. We shall also consider the need for specialised formalisms for describing clinical processes, drawing on experience with workflow languages (e.g. BPEL4WS, BPMN), guideline modelling languages (e.g. GLIF, PROforma), AI planning languages (e.g. PDDL, OCL) and “agent” programming systems (e.g. LALO, 3APL). The adoption of clinical workflow technology will greatly benefit from the availability of appropriate languages for declaratively representing processes of care. To explore some of the requirements for a clinical workflow technology we will review and critique the PROforma process modelling language and the Arezzo® and Tallis toolsets which use it. The discussion will be illustrated with deployed applications and operational prototypes. 5th International Conference on BPM 2007 Workshop: BPM in Healthcare

Traditional workflow management systems (WFMSs) are not flexible enough to support loosely-structured processes. Furthermore, flexibility in contemporary WFMSs usually comes at a certain cost, such as lack of support for users, lack of methods for model analysis, lack of methods for analysis of past executions, etc. DECLARE is a proto-type of a WFMS that uses a constraint-based process modeling language for the development of declarative models describing loosely-structured processes. In this paper we show how DECLARE can support loosely-structured processes without sacrificing important WFMSs features like user support, model verification, analysis of past executions, changing models at run-time, etc.