Business Processes Contextualisation via Context Analysis



Similar documents
Goal-Based Self-Contextualization

Informe Técnico / Technical Report

Aligning Software Configuration with Business and IT Context

Transforming Socio-Technical Security Requirements in SecBPMN Security Policies

Business Process Configuration with NFRs and Context-Awareness

Business Process Models andVariations in Variability

A Holistic Approach to Security Attack Modeling and Analysis

Location-based Software Modeling and Analysis: Tropos-based Approach

Capturing Variability in Business Process Models: The Provop Approach

Keywords: Forward Logistics, Reverse Logistics, Supply Chain Management, Mathematical Programming, Beverage Industry.

Ulmer Informatik-Berichte. Dealing with Variability in Process-Aware Information Systems: Language Requirements, Features, and Existing Proposals

The Design of an Agent-Based Production Scheduling Software Framework for Improving Planning-Scheduling Collaboration

Prediction of Business Process Model Quality based on Structural Metrics

A Model-driven Approach to Flexible Multi-Level Customization of SaaS Applications

Enumerating possible Sudoku grids

A Variability Viewpoint for Enterprise Software Systems

DESIGNING A VOIP BASED LANGUAGE TEST

STS-Tool: Security Requirements Engineering for Socio-Technical Systems

Towards a Comprehensive Design-time Compliance Management: A Roadmap

Informe Técnico / Technical Report

Requirements Engineering for Social Applications

The Emerging ISO International Standard for Certification of Software Engineering Professionals

A Model-Driven Approach for Developing Self-Adaptive Pervasive Systems

CRISTAL: Collection of Resource-centrIc Supporting Tools And Languages

Context-driven Business Process Modelling

Learning analytics in the LMS: Using browser extensions to embed visualizations into a Learning Management System

Process Change Patterns: Recent Research, Use Cases, Research Directions

Business Process Measurement in small enterprises after the installation of an ERP software.

From Workflow Design Patterns to Logical Specifications

Structural Detection of Deadlocks in Business Process Models

The epistemic structure of de Finetti s betting problem

DIAMANT RUBEN GONZALEZ-RUBIO DÉPARTEMENT DE GÉNIE ÉLECTRIQUE ET DE GÉNIE INFORMATIQUE UNIVERSITÉ DE SHERBROOKE

How Stage Gate Process Supports CbC: Case Study

IMPROVING THE DESIGN-CONSTRUCTION INTERFACE

6/21/2013. High Level Overview

MATHEMATICS KNOWLEDGE FOR TEACHING WITHIN A FUNCTIONAL PERSPECTIVE OF PRESERVICE TEACHER TRAINING

New Regulation for Doctoral Studies in Spain

A System for Interactive Authorization for Business Processes for Web Services

SQLMutation: A tool to generate mutants of SQL database queries

"34".!! "#$%&'()!*+!(#!*,!-%!.%/01%$2&%!

Policy Modeling and Compliance Verification in Enterprise Software Systems: a Survey

pm4dev, 2015 management for development series Project Schedule Management PROJECT MANAGEMENT FOR DEVELOPMENT ORGANIZATIONS

Formal Engineering for Industrial Software Development

Managing and Tracing the Traversal of Process Clouds with Templates, Agendas and Artifacts

OpenText Content Server: Automated Routing and Coding (ARC)

Orchestrating an SOA with Rules

Event correlation for process discovery from web service interaction logs

InvGen: An Efficient Invariant Generator

A Semantical Perspective on Verification of Knowledge

An Experiment on the Effect of Design Recording on Impact Analysis

A Coordination Protocol for Higher Education Degrees

Improving Software Engineering Practice with HCI Aspects


PREDICTING SUCCESS IN THE COMPUTER SCIENCE DEGREE USING ROC ANALYSIS

Coverability for Parallel Programs

GENERAL CONTRACTING CONDITIONS

The theory of the six stages of learning with integers (Published in Mathematics in Schools, Volume 29, Number 2, March 2000) Stage 1

Copyright Soleran, Inc. esalestrack On-Demand CRM. Trademarks and all rights reserved. esalestrack is a Soleran product Privacy Statement

Exam Introduction Mathematical Finance and Insurance

Likewise, we have contradictions: formulas that can only be false, e.g. (p p).

Linking BPMN, ArchiMate, and BWW: Perfect Match for Complete and Lawful Business Process Models?

Acceptance of reservations for a rent-a-car company J.Alberto Conejero 1, Cristina Jordan 2, and Esther Sanabria-Codesal 1

We would like to state the following system of natural deduction rules preserving falsity:

An Ontology Based Method to Solve Query Identifier Heterogeneity in Post- Genomic Clinical Trials

A Multi-agent System for Knowledge Management based on the Implicit Culture Framework

CHAPTER 7 GENERAL PROOF SYSTEMS

A Systematic Security Approach in Software Requirements Engineering

Disjunction of Non-Binary and Numeric Constraint Satisfaction Problems

Amortized Loan Example

Development of a tool for the evaluation and improvement of the energy management in small and medium enterprises (SMEs)

Social Team Characteristics and Architectural Decisions: a Goal-oriented Approach

Ulmer Informatik-Berichte. Correct Configuration of Process Variants in Provop. Alena Hallerbach, Thomas Bauer, Manfred Reichert

Mining Configurable Process Models from Collections of Event Logs

To introduce software process models To describe three generic process models and when they may be used

Introduction to Software Verification

Textkernel Search! User Guide. Version , Textkernel BV, all rights reserved

market research p&a international Market Plaza Esomar: Global and Local Partner in Market Research

Distribution Training Guide. D110 Sales Order Management: Basic

Turning Emergency Plans into Executable

Unit 7 The Number System: Multiplying and Dividing Integers

CSL105: Discrete Mathematical Structures. Ragesh Jaiswal, CSE, IIT Delhi

An Aspect-Oriented Product Line Framework to Support the Development of Software Product Lines of Web Applications

MULTI-CRITERIA OPTIMIZATION IN WORKFORCE MANAGEMENT

The tool you need to effectively measure and report healthcare marketing ROI!

On the general structure of ontologies of instructional models

Model Checking: An Introduction

Trust but Verify: Authorization for Web Services. The University of Vermont

Application Notes for Configuring Dorado Software Redcell Enterprise Bundle using SNMP with Avaya Communication Manager - Issue 1.

Software Engineering. Software Processes. Based on Software Engineering, 7 th Edition by Ian Sommerville

Reasoning-based Techniques for Dealing with Incomplete Business Process Execution Traces

An Empirical Study of Application of Data Mining Techniques in Library System

Workflow Automation and Management Services in Web 2.0: An Object-Based Approach to Distributed Workflow Enactment

XBRL Interoperability through a Multidimensional Data Model

EDIminer: A Toolset for Process Mining from EDI Messages

Nan Kong, Andrew J. Schaefer. Department of Industrial Engineering, Univeristy of Pittsburgh, PA 15261, USA

Trading Rules of the Georgian Stock Exchange

Alternative Job-Shop Scheduling For Proton Therapy

Principles of Software Engineering: Software Methodologies. COSI 120b, Spring 2005

Incorporating Internal Gradient and Restricted Diffusion Effects in Nuclear Magnetic Resonance Log Interpretation

A Propositional Dynamic Logic for CCS Programs

Transcription:

Business Processes Contextualisation via Context Analysis Jose Luis de la Vara 1, Raian Ali 2, Fabiano Dalpiaz 2, Juan Sánchez 1, and Paolo Giorgini 2 1 Centro de Investigación en Métodos de Producción de Software Universidad Politécnica de Valencia, Spain {jdelavara, jsanchez}@pros.upv.es 2 Department of Information Engineering and Computer Science University of Trento, Italy {raian.ali, fabiano.dalpiaz, paolo.giorgini}@disi.unitn.it Abstract. Context-awareness has emerged as a new perspective for business process modelling. Even though some works have studied it, many challenges have not been addressed yet. There is a clear need for approaches that (i) facilitate the identification of the context properties that influence a business process and (ii) provide guidance for correct modelling of contextualised business processes. This paper addresses this need by defining an approach for business process contextualisation via context analysis, a technique that supports reasoning about context and discovery of its relevant properties. The approach facilitates adequate specification of context variants and of business process execution for them. As a result, we obtain business processes that fit their context and are correct. Keywords: business process modelling, context-awareness, business process contextualisation, context analysis, correctness of business process models. 1 Introduction Traditional approaches for business process modelling have not paid much attention to the dynamism of the environment of a business process. However, business processes are executed in an environment in which changes are usual, and modelling perspectives that aim to represent and understand them are necessary. Context-awareness has recently appeared as a new perspective for business process modelling to meet this need [3]. It is expected to improve business process modelling by explicitly addressing fitness between business processes and their context. The context of a business process is the set of environmental properties that affect business process execution. Therefore, these properties should be taken into account when designing a business process. If context is analysed when modelling a business process, then identification of all its variants (relevant states of the world in which the business process is executed) and definition of how the business process should be executed in them are facilitated.

Some works have contributed to the advance of context-aware business process modelling by addressing issues such as context-aware workflows [4], general principles (e.g. [3]) and modelling of context effect (e.g. [2]). However, research on this topic is still at an initial stage and many challenges have not been addressed yet. This paper aims to advance in research on context-aware business process modelling by dealing with two of these challenges: 1) provision of techniques for determination of the relevant context properties that influence a business process, and; 2) provision of mechanisms and guidance for correct business process contextualisation. The objectives of the paper are to determine how business process context can be analysed, how it can influence business processes, how to create contextualised business process models, and how to guarantee their correctness. These objectives are achieved by defining an approach for business process contextualisation via context analysis [1], which is a technique that aims to support reasoning about context and discovery of contextual information to observe. Context analysis is adapted in the paper for analysis of business process context. The approach provides mechanisms and guidance that can help process designers to reason about business process context and to model business processes that fit their context and are correct. Context properties and variants are analysed in order to determine how they influence a business process, to guarantee that a business process is properly executed in all its context variants, and to correctly model contextualised business processes. The next sections present the approach and our conclusions, respectively. 2 Approach Description The approach consists of four stages (Fig. 1): modelling of initial business process, analysis of business process context, analysis of context variants and modelling of contextualised business process. First, an initial version of the business process that needs to fit its context is modelled. Next, the rest of stages have to be carried out while relevant context variations (changes) are found and they are not represented in the business process model. Relevant context variations influence the business process and imply that business process execution has to change. If a context variation is found, then business process context is analysed to find the context properties that allow process participants to know if a context variant holds. A context analysis model is created, and context variants of the business process are then analysed. Finally, a contextualised business process model is created on the basis of the final context variants and their effect on the business process. Fig. 1. Business process contextualisation

As a running example, product promotion in a department store is used (Fig. 2). The business process has been modelled with BPMN, and it does not reflect context variations such as the fact that customers do not like being addressed if they are in a hurry. The paper focuses on contextualisation of the task Find potential buyer. Fig. 2. Initial business process model 2.1 Analysis of Business Process Context Business process context is analysed in the second stage of the approach. This stage aims to understand context, to reason about it and to discover the context properties that influence a business process. For these purposes, context analysis (which has been presented in the requirements engineering field) has been adapted for analysis of business process context. Further details about context analysis can be found in [1]. Context is specified as a formula of world predicates, which can be combined conjunctively and disjunctively. World predicates can be facts (they can be verified by a process participant) or statements (they cannot be). The truth value of a statement can be assumed if there is enough evidence to support it. Such evidence comes from another formula of world predicates that holds. A context is judgeable if there exists a formula of facts that supports it, and thus implies it. Identifying a judgeable context can be considered the main purpose of this stage. The facts of the formula correspond to the context properties that characterise the context and its variants, and their truth values influence business process execution. A context analysis model (Fig. 3) is created to facilitate reasoning about business process context and discovery of the facts of the formula that implies it. C1: A potential buyer is found F1: Customer [c] asks for product [p] S1: [c] may accept being addressed for promotion [pr] of [p] F2: [c] asks about [pr] S2: [c] is not in a hurry F3: [c] is near S3: [c] may be interested in [pr] S4: [c] does not have to work F4: [c] walks slowly S5: [c] is just examining products [p ] F5: [c] has not bought [p] S6: [c] may want to buy [p] F6: It is holiday in the region [r] of the department store [ds] F7: [c] has been just looking at [p ] for a while F8: [c] has not bought any product [p ] F9: [c] is looking at [p] S7: [c] is interested in a product [p ] that is related to [p] Legend Statement Fact And Or Support Imply Fig. 3. Context analysis model F10: [c] has bought [p ] F11: [c] asks for [p ]

2.2 Analysis of Context Variants The main purposes of this stage are to adequately define the (final) context variants of a business process and that they allow correct business process contextualisation. A context variant corresponds to a set of facts whose conjunction implies a context. Fig. 4 shows the eleven initial context variants for C1, which is analysed in Fig. 1. Initial Context Variants Final Context Variants {F1} CV1: {F1} {F2} CV2: {F2} {F3, F4, F5, F9} CV3: {F3 (F4, F5, F9)} {F3, F4, F5, F10} CV4: {F3 (F4, F5, F10)} {F3, F4, F5, F11} CV7: {F4, F11 F5} {F3, F5, F6, F9} CV5: {F6 F3 (F5, F9)} {F3, F5, F6, F10} CV6: {F6 F3 (F5, F10)} {F3, F5, F6, F11} CV8: {F6 F11 F5} {F3, F5, F7, F8, F9} CV9: {F3 (F7, F8, F9)} {F3, F5, F7, F8, F10} CV10: {F7, F11 F8} {F3, F5, F7, F8, F11} Fig. 4. Context variants Correctness of business processes is usually related to its soundness [5]. For business process executions that are defined from context variants, two situations can impede soundness of a contextualised business process. The first one is that a context variant contains conflicting facts. The second situation is to follow a sequence of fact verifications that will not allow a business process instance to be finished. These situations are avoided by analysing the context variants. For this purpose, a table is created to specify the relationships between facts. The table also aims to obtain context variants whose sets of facts are the minimum ones. An example is shown in Table 1, which specifies the relationships between the facts of the initial context variants of Fig. 4. The relationships are specified as follows. Given a pair of facts Fr (fact of a row) and Fc (fact of a column), their relationship can be: X (no context variant contains Fr and Fc together); Pr (Fr verification will precede Fc verification); Pc (opposite to Pr ); Kr (Fr truth value will be known before Fc verification); Kc (opposite to Kr ); Ur (Fr is always true when the Fc is true, thus Fr verification will be unnecessary when Fc is true); Uc (opposite to Ur ); C (Fr and Fc are conflicting); - (no relationship exists). Finally, context variants are refined by specifying sequence of fact verification ( ) and removing conflicting variants and unnecessary facts (Fig. 4). Table 1. Relationships between facts F11 F10 F9 F8 F7 F6 F5 F4 F3 F2 F1 X X X X X X X X X X F2 X X X X X X X X X F3 Ur Pr Pr Pr Pr Kc Pr Pr F4 - - - X X X - F5 Pc - - Ur - Kc F6 Kr Kr Kr X X F7 - - - - F8 Pc C - F9 X X F10 X

2.3 Modelling of Contextualised Business Process A contextualised business process is modelled on the basis of its final context variants. The first step is determination of the tasks that will be part of the business process. They can correspond to: 1) tasks of the initial business process model that are not influenced by context); 2) tasks that are defined from refinement of the tasks of the initial business process model (e.g. Address customer refines Find potential buyer ), and; 3) tasks that make facts true (e.g. Approach customer makes F3 true). If a task of the latter type is executed when a given fact is false, then the fact turns into true. These facts are called manageable. Table 2. Relationships between tasks and facts F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 T1: Approach customer U U M - - - - - - - U T2: Address customer U U Sc Sc1 Sc1 Sc1 Sc1 Sc1 Sc1 Sc1 U Once tasks are determined, a table is created to specify their relationships with the facts of the final context variants. An example is shown in Table 2. The relationships are specified as follows. Given a fact F, a set of facts φ and a task T, their relationship can be: M (T allows F to be manageable); U (T execution will be unnecessary if F is true); Sc (T execution will succeed F verification); ScX (where X is a number; T execution will succeed verification of the facts of φ); - : (no relationship exists) CE1: F1 CE2: F2 CE3: (F3 T1) (F4, F5, F9) T2 CE4: (F3 T1) (F4, F5, F10) T2 CE5: F6 (F3 T1) (F5, F9) T2 Fig. 5. Contextualised executions CE6: F6 (F3 T1) (F5, F10) T2 CE7: F4, F11 F5 CE8: F6 F11 F5 CE9: (F3 T1) (F7, F8, F9) T2 CE10: F7, F11 F8 The next step for modelling of a contextualised business process is specification of its contextualised executions (Fig. 5). A contextualised execution is a set of fact verifications and task executions that specifies a correct execution of a business process or of a fragment of a business process for a context variant. Assistant Fig. 6. Contextualised business process model

Contextualised executions are specified by extending the final context variants of a business process with the execution sequence of its tasks ( ). The manageable facts and their associated tasks are put in brackets and the symbol is put between them: either the fact is true or the task has to be executed. Finally, a contextualised business process model is created on the basis of the constraints (fact verification and task execution sequences) that the contextualised executions impose. BPMN has been extended by labelling its sequence flows for specification of formulas that have to hold so that a sequence flow is executed. Fact and formula verification is represented by means of gateways. Fig. 6 shows the effect of contextualisation of the task Find potential buyer for the running example. 3 Conclusions and Future Work This paper has addressed several challenges of context-aware business process modelling in order to allow research on it to further advance. As a result, an approach for business process contextualisation has been presented. The approach adapts context analysis for analysis of business process context, and provides mechanisms and guidance for analysis of business process context and its variants and for modelling of contextualised business processes. It facilitates discovery and adequate specification of relevant context properties in the form of facts, as well as of the relationships between facts and between facts and tasks of a contextualised business process. These relationships affect business process execution. Furthermore, the mechanisms and guidance can guarantee that a contextualised business process fits its context and is sound. As future work, we have to address approach automation and formal evaluation. Acknowledgements. This work has been developed with the support of the Spanish Government under the projects SESAMO TIN2007-62894 and HI2008-0190 and the program FPU AP2006-02324, partially funded by the EU Commission through the projects COMPAS, NESSOS and ANIKETOS, and co-financed by FEDER. The authors would also like to thank Amit K. Chopra for his useful comments. References 1. Ali, R., Dalpiaz, F., Giorgini, P.: A Goal-based Framework for Contextual Requirements Modeling and Analysis. Requirements Engineering Journal (to appear) (2010) 2. Hallerbach, A., Bauer, T., Reichert, M.: Capturing Variability in Business Process Models: The Provop Approach. Journal of Software Maintenance and Evolution (to appear) (2010) 3. Rosemann, M., Recker, J., Flender, C.: Contextualisation of business processes. International Journal of Business Process Integration and Management 3(1):47-60 (2008) 4. Smanchat, S., Ling, S., Indrawan, M.: A Survey on Context-Aware Workflow Adaptations. In: MoMM2008, pp. 414-417 (2008) 5. Weske, M.: Business Process Management. Springer (2007)