The tutorial will cover the principles and the normative quality characteristics as well as the standardized procedures of information quality assurance resp. software system quality assurance (comprising V&V, test, measurement and assessment) for procedural, object-oriented, and agent-based dependable software systems. Attendees will exercise proven techniques for goal-directed measurement, scaling and assessment for software certification. Assessment of both the software product as well as the software process will be discussed with respect to its relevance for such acceptance assessments. A standardized process model for measurement, assessment and certification of dependable software will be used to make the attendees familiar with this comprehensive assessment procedure and to learn how to embed it into today's standardized or non-standardized software processes.
Basic knowledge in mathematics and some knowledge of software methods and tools are required. Emphasis will be given to selected advanced topics depending on the needs of participants.
Overall aim of the tutorial is to make the attendees aware of the development of the technology and to provide practical approaches to the quality problem. Putting together the results and findings of several projects on software quality and productivity the tutorial will especially address the needs of software managers and software developers.
1. General overview: What,
where and how? Tutorial organization.
2. Introduction: Refresh of state of the art and practice in industry. The Q question. Information quality, software quality, and data quality. Norms, standards, and terminology.
3. Software process and software product: Definition and representation of development processes. Definition and representation of software products, information systems as products.
4. Quality and productivity modeling: Test, verification and validation with respect to evaluation and certification. Characteristics, metrics and assessment.
5. Process evaluation and certification: Capability assessment procedures. Process assessment norms and standards (1- CMM, CMMI, IS09000, IS012207, IS015288, IS015704; 2- TicklT, Trillium, ami, SPICE
6. Product evaluation and certification: Motivation and justification of evaluation and certification. The tailored evaluation procedure (1- Evaluation of bespoken systems according IS09126 and ISO 14598 and the new ISO 25000 series; 2- Evaluation of COTS according ISO 12119)
7. Summary and conclusion: What did we hear and why? What do we expect from the big Q discussion? Concluding reflections.
Lecturer: A senior scientist at Fraunhofer Institute in Sankt Augustin (Germany), Hans-Ludwig Hausen, has more than 25 years of practical experience in computer-aided software engineering, software quality insurance, software process modeling and tailoring in the context of some thirty large software engineering projects for governmental agencies and industry.
This tutorial covers the theory of requirements engineering for software systems. It consists of simplified techniques applied to case studies. The goal is to give insight to processes and techniques used to prepare industrial software requirements. Problem solving is illustrated with trade-off analysis. Establishing a Measurable Operational Value from a customer prospectus, establishing quantitative requirements feature sets, setting priorities using a simplified quality function deployment approach, validating the feature packages with prototypes, and extending the prototypes to models are specific topics covered. The tutorial includes ways to estimate staffing, schedules and reliability and evaluating the resulting product with ICED-T metrics.
One of the case studies illustrates opportunity analysis, customer value proposition, and requirements modeling and analysis. The following textbook will be used: “Trustworthy Systems Through Quantitative Software Engineering,” Lawrence Bernstein and C. M. Yuhas, Wiley, 2005, ISBN 0-471-69691-9
Topics that will be covered:
1.1 Requirements Analysis
1.2 Requirements Process
1.4 Systems Engineering Methodology
1.5 Software as a Business
2. Requirements Engineering
2.1 Requirements Context
2.2 Requirements Elicitation
2.4 Quality Function Deployment (QFD)
2.5 Case Study
3. Opportunity Analysis
3.1 Introduction to Function
3.2 Lambda Protocol
3.4 Cost Estimation
4. Consequences of Poor Requirements
4.1 Measurable Operational
4.2 Feature Creep
4.3 Market Miss
4.4 Case History: London Ambulance Dispatch
Typical Case History: Your company is hired by a new bed and breakfast chain in Indiana to build a software system. There are four B&Bs in the chain. They have five bedrooms for guests. They want a system to manage the reservations and the arrival times for guests. Checkout is noon and check-in is two o’clock. Guests may negotiate later checkout times. They wish to track gross revenues. They accept credit cards. They negotiate room price and one-night deposit with the customer. They hold reservations without deposits for one month. Design use cases for the system, an MOV and a simplified QFD.
Lecturer: Lawrence Bernstein is a recognized expert in software engineering. He is teaching at Stevens Institute of Technology, where he directs the Stevens Quantitative Software Engineering Program. He had a 35-year distinguished career at Bell Laboratories in managing large software projects.
This tutorial is proposed by members of the MeFoSyLoMa (Méthodes Formelles pour les Systèmes Logiciels et Matériels), a working group that gathers teams from several academic organizations in the Paris area: Université Paris-Dauphine (LAMSADE Lab), Université P. & M. Curie (LIP6 Lab), Université Paris 13 (LIPN Lab), ENST (LTCI Lab), and CNAM (CEDRIC Lab).
These teams have extensive
knowledge and experience in the design, analysis, and implementation
of distributed systems. The cooperation within the group aims at
joining forces, sharing experiences, and building joint projects to
solve issues in the design of reliable distributed systems. The
tutorial consists of three parts:
Part I: Dedicated specification languages and models
Many sorts of models can be used to specify a complex system. Hence, it might prove difficult, for a non-experienced designer, to choose among this large collection. Therefore, the first part of he presentation is dedicated to criteria that should be taken into account before choosing a modeling formalism. The second part of the presentation is concerned with guidelines to start writing the detailed specification, considering data types structures, simple dynamic systems, and dynamic systems structured using subsystems.
Lecturers: Laure Petrucci and Christine Choppy, Professors at University Paris 13, and members of the CNRS LIPN Lab.
Part II: Dedicated verification techniques
The diversity of verification methods may puzzle the engineer faced to the choice of the appropriate technique for analyzing her/his system. The first part of the presentation aims at clarifying the bases of such a choice by discussing three critical questions associated with verification. The second part of the presentation details one of the most successful verification methods in order to tackle the increasing complexity of systems: the decision diagram-based (BDD) methods.
Lecturers: Serge Haddad, Professor at University Paris-Dauphine, member of the CNRS Lamsade Lab, and Yann Thierry-Mieg, Associate Professor at University P. & M. Curie, member of the CNRS LIP6 lab.
Part III: Application to distributed systems
This part is dedicated to the effective use of the techniques presented in the design and implementation of real systems. First, the PolyORB experience is presented. PolyORB is a middleware dedicated to distributed real-time systems. It thus requires high reliability that is achieved by means of an original architecture on which formal verification of qualitative properties. Secondly, the talk deals with the design of adaptive reactive systems, i.e. systems that dynamically adapt their architecture depending on the context of the execution. We use the formalism of timed automata for the design of the module behavior. Hence, it is possible to evaluate beforehand the properties of the system (regarding logical correctness and timeliness), thanks to model-checking and simulation techniques.
Lecturers: Fabrice Kordon, Professor at University P & M. Curie and head of the MoVe (Modeling and Verification) team in the CNRS LIP6 Lab, Hanna Klaudel, Professor at University Évry-Val d’Essonne and head of the LIS (Languages, Interaction, Simulation) team in the CNRS IBISC Lab.
Business Process Management (BPM) is very popular today in both business and IT circles. At one level, BPM emphasizes creating enterprise architectures that show how IT resources can be used to support business goals and initiatives. In effect is creates a better way to facilitate communication between business and IT managers. At another level BPM focuses redesigning business processes to meet the challenges that companies face today – problems resulting from globalization, outsourcing, mergers and the rise of virtual business processes that spread across multiple companies and countries. BPM entails the coordination of diverse groups to redesign business processes and assure improved business performance. Popular BPM Software products can be used to create and then dynamically control and monitor the use of human (workflow) and IT resources. Many BPM products focus on reorganizing and improving the way companies use ERP and CRM applications. Other BPMS applications emphasize a Service-Oriented Architecture (SOA) and, together, BPM and SOA are driving a number of standards initiatives ranging from ebXML and BPEL to BPMN and the MDA Process and Rules metamodels.
This tutorial will cover the development and current emphasis of those involved in Business Process Management, with a special emphasis on modeling and IT. We will consider all of the different aspects of today’s Business Process Management movement, providing an overview and context for understand how these initiatives will develop in the next few years.
We will consider examples of companies that have made heavy commitments to BPM and the results they have achieved.
1. A Broad Overview of
Business Process Management
Definitions and Data on the Size and Direction of the Market
2. A Look at BPM Enterprise Level Initiatives
Business Process and Enterprise Architectures
SCOR and the use of OR Frameworks as Management Tools
Case Study: SCOR, Compaq and HP
Changing Companies to Emphasize Process Management
Case Study: Boeing GMS
3. A Look at Process Level Initiatives
Process Redesign and Improvement
Modeling Standards, Metamodels and BP Modeling Tools
BPM and Six Sigma
4. A Look at Implementation Level Initiatives
BPM Systems Tools and Applications
BAM, Performance Monitoring, BI and Dashboards
BPM Systems and ERP and CRM Applications
Case Study: Air Products
BPM, SOA and Evolving Standards
5. What Will Happen in the Next Few Years?
Lecturer: Paul Harmon is the author of Business Process Change, the best selling BPM book in the US. He is the executive editor of Business Process Trends : a portal that business and IT managers use to keep up on BP developments –, and he is a widely respected consultant and speaker on business process management topics
Copyright (C) CNAM-CMSL 2006, Tous droits réservés.