Prof.Dr. Sylvie Trudel, Université du Québec à Montréal (UQAM)
Sylvie Trudel has worked since 1985 in software development and software process improvement in various fields such as aerospace, insurance, banking, scientific sector and public and private organizations. After completing her Ph.D. at ETS in 2012, she became a professor of software engineering at UQAM. She has applied the COSMIC method in the industry since 2003 to build estimation models, measure the software process productivity, and benchmark organizational performances.
She teaches now the COSMIC method in software project management and software measurement courses, in addition to mentoring students using COSMIC as part of their master degree project. She became an agile coach after her Scrum Master certification in May 2006 and is co-author of the book Choosing Agility: From Software Development to Governance (in French, Dunod [France], 2011). She has also been a lecturer at the University of Sherbrooke, where she taught the software measurement course to software engineering graduate students, including functional size measurement with COSMIC method.
Software Measurement Fraud Detection: Concepts and Solutions
Measuring attributes of software activities and artefacts is generally applied to monitor achievement of objectives, such as delivering the defined scope within budget and schedule, with the expected or stated level of quality, while keeping the customer and the staff satisfied. As it may seems natural that values resulting from measurement activities reflect the facts, there are a number of cases where facts might have been “twisted” or “distorted”, intentionally or not, leading to doubtful results that would be inaccurate, incorrect, or even fraudulent: these may then be used as a basis for decision making by managers who have no clue that these results are wrong despite verification mechanisms applied upon them. What would be considered as a software measurement fraud or malpractice? And, in such cases, how can we detect software measurement frauds and malpractices? Once a fraud or malpractice is detected, what can be done to correct the results and prevent frauds or malpractices from perpetuating? This presentation is exploring some of the symptoms and the causes of software measurement frauds and malpractices, which are either intellectual or practical, and solutions are proposed to overcome these problems, whether they happen in a traditional or Agile context.