Arjan J. Mooij

Service Interaction: Patterns, Formalization, and Analysis

As systems become more service oriented and processes increasingly cross organizational boundaries, interaction becomes more important. New technologies support the development of such systems. However, the paradigm shift towards service orientation, requires a fundamentally different way of looking at processes. This survey aims to provide some foundational notions related to service interaction. A set of service interaction patterns is given to illustrate the challenges in this domain. Moreover, key results are given for three of these challenges: (1) How to expose a service?, (2) How to replace and refine services?, and (3) How to generate service adapters? These challenges will be addressed in a Petri net setting. However, the results extend to other languages used in this domain.

User-guided discovery of declarative process models

Process mining techniques can be used to effectively discover process models from logs with example behaviour. Cross-correlating a discovered model with information in the log can be used to improve the underlying process. However, existing process discovery techniques have two important drawbacks. The produced models tend to be large and complex, especially in flexible environments where process executions involve multiple alternatives. This “overload” of information is caused by the fact that traditional discovery techniques construct procedural models explicitly showing all possible behaviours. Moreover, existing techniques offer limited possibilities to guide the mining process towards specific properties of interest. These problems can be solved by discovering declarative models. Using a declarative model, the discovered process behaviour is described as a (compact) set of rules. Moreover, the discovery of such models can easily be guided in terms of rule templates. This paper uses DECLARE, a declarative language that provides more flexibility than conventional procedural notations such as BPMN, Petri nets, UML ADs, EPCs and BPEL. We present an approach to automatically discover DECLARE models. This has been implemented in the process mining tool ProM. Our approach and toolset have been applied to a case study provided by the company Thales in the domain of maritime safety and security.

Reducing Adapter Synthesis to Controller Synthesis

Service-oriented computing aims to create complex systems by composing less-complex systems, called services. Since services can be developed independently, the integration of services requires an adaptation mechanism for bridging any incompatibilities. Behavioral adapters aim to adjust the communication between some services to be composed in order to establish proper interaction between them. We present a novel approach for specifying such adapters, based on domain-specific transformation rules that reflect the elementary operations that adapters can perform. We also present a novel way to synthesize complex adapters that adhere to these rules, viz., by consistently separating data and control, and by using existing controller-synthesis algorithms. Our approach has been implemented, and we discuss some example applications, including real business processes in WS-BPEL.

Relating fair testing and accordance for service replaceability

Abstract The accordance pre-order describes whether a service can safely be replaced by another service. That is, all partners for the original service should be partners for the new service. Partners for a service interact with the service in such a way that always a certain common goal can be reached. We relate the accordance pre-order to the pre-orders known from the linear–branching time spectrum, notably fair testing. The differences between accordance and fair testing include the modeling of termination and success, and the parts of the services that cannot be used reliably by any partner. Apart from the theoretical results, we address the practical relevance of the introduced concepts.

Non-local choice and beyond: intricacies of MSC choice nodes

MSC is a visual formalism for specifying the behavior of systems. To obtain implementations for individual processes, the MSC choice construction poses fundamental problems. The best-studied cause is non-local choice, which e.g. is unavoidable in systems with autonomous processes. In this paper we characterize two additional problematic classes of choice nodes. Based on these three classes we point out some errors in related work. Extending our work on pragmatic implementations of non-local choice, we motivate a different choice semantics which allows a little more behavior. Finally, inspired by practical case studies, we present the first implementation approach for non-local choice nodes that can handle arbitrary numbers of processes.

Proof Techniques for Adapter Generation

We study the composition and substitution of services from a theoretical perspective. An important notion is the operating guideline of a service y , which is defined as the set of services x such that the result of connecting x and y has a certain desired property. We define several related notions and derive results for them in a general context, thus abstracting from the underlying formalism, be it process algebra, Petri nets or something else. We then focus on the open Petri-net (and oWFN) formalism, and address the automated generation of adapters.

Progress in deriving concurrent programs: emphasizing the role of stable guards

We present some techniques to obtain smooth derivations of concurrent programs that address both safety and progress in a formal manner. Our techniques form an extension to the calculational method of Feijen and van Gasteren using a UNITY style progress logic. We stress the role of stable guards, and we illustrate the derivation techniques on some examples in which progress plays an essential role.

Analyzing Vessel Behavior Using Process Mining

In the maritime domain, electronic sensors such as AIS receivers and radars collect large amounts of data about the vessels in a certain geographical area. We investigate the use of process mining techniques for analyzing the behavior of the vessels based on these data. In the context of maritime safety and security, the goal is to support operators in identifying suspicious behavior that may indicate accidents or undesired activities such as smuggling and piracy. Our approach consists of two phases. In the first phase, process mining is used offline to extract from historical data a reference model of the normal vessel behavior, which can be adapted by experienced operators and domain experts. In the second phase, process mining is used online to verify whether the current vessel behavior is compliant with the reference model, thus allowing for the identification of suspicious behavior.

Early fault detection in industry using models at various abstraction levels

Most formal models that are used in the industry are close to the level of code, and often ready to be used for code generation. Formal models can also be analysed and verified in order to detect any faults. As the first formal models are often such code-level models, their analysis not only reveals a lot of detailed design faults, but also the more relevant conceptual faults in the design and the requirements. Our observations are based on our experiences in an industrial development project that uses a commercial tool for formal modelling, compositional verification, and code generation. In addition to the provided tool functionality, we have introduced formal techniques to detect conceptual faults during the earlier design and requirements phases. To this end we have made additional formal models, both for the requirements and for the early designs at various abstraction levels. We have analysed these models using simulation and interactive visualization, and we have compared them using refinement checking.

Soundness-preserving refinements of service compositions

Soundness is one of the well-studied properties of processes; it denotes that a final state can be reached from every state that is reachable from the initial state. Soundness-preserving refinements are important for enabling the compositional design of systems. In this paper we concentrate on refinements of service compositions. We model service compositions using Petri nets, and consider specific pairs of places that belong to different services. Starting from a sound service composition, we show how to check whether such a pair of places can be refined by another sound service composition, so that soundness is preserved through the refinement.