|
Measurement-Based Software Project Management
You can't control what you can't measure. -- Tom DeMarco, Controlling Software Projects. The best criteria for the value of a metric is the degree to which it helps us make a decision. -- Dr. Barry W. Boehm, TRW and UCLA. What is not measurable, make measurable. -- Galileo Galilei 1564-1642. Software metrics will help you develop better software in your organization. Some metrics will be more useful to your specific environment than others; some, if analyzed and used properly, will result in immediate pay-back, others may need to be collected over a period of time before predictable trends develop. Course Description Without measurement, we can’t possibly know how we are progressing. Are our estimates getting more accurate? Are our defects caught earlier and earlier in the life cycle? Are we putting fewer defects into our products? Are our process improvements making a difference? Without measurement, software is destined to remain an art form or craft -- not something today's systems can afford. Quantification of productivity and quality, and the factors that influence them, have practical value to both the management and the technical community within the software industry. Measurement, a critical role in effective software development, provides a scientific basis for software engineering to be a true engineering discipline. During this course, the student learns to recognize the power in exploiting "lifecycle-step-wise quality measurements" to predict, manage and/or control software quality. Upon completion of this course, the participant should be able to: • Explain the value of keeping software-related metrics. • List categories of metrics and give examples of each. • Describe where metrics occur in the software life cycle. • Identify a “good” metric. • Describe a “starter set” of practical software metrics, including quality-related measures. • Explain the basic concepts of setting goals for metrics. • Build a model for measuring the value of early defect detection and removal. • Apply “planning coefficients” to be used in future project cost, sizing, and schedule estimation.
Drawing upon recognized experts in the field, the participant will experience the following in class: The activities begin with a literature search to compile typical industry data for productivity, expected defect generation, and detection/removal rates at each major lifecycle step in the development process. Then, using such typical data, the student builds a software quality model in the form of a "quality synthesis table" in which the quality of a theoretical software project is measured at each successive "synthesis" step. By making small changes to the detection values used in the resultant quality model, the student sees the value/cost of removal at each step, and learns how minor process improvements in progressively earlier defect removal activities have dramatic positive effects on the resulting quality, cost, speed to market, and field maintenance requirements. The student then replaces the industry data with empirical (actual) project-specific data (obtained as a part of the course) and measures/determines the influence coefficients of quality improvement at each synthesis (lifecycle) step, and calculates the resultant impact on the total effective project cost and the associated quality/reliability risk. Finally, the model is used to extract "planning coefficients" to be used in future project cost, sizing, and schedule estimation. This course will cover: Definitions -- What is a Software Metric? Types of Software Metrics. What may be measured. Measurement scales. The importance of metrics. What Makes a “Good” Metric? Usable metrics. Goal, Question, Metric paradigm. Metrics analysis. Stakeholder feedback. Combining metrics. Industry standard software metrics. Defect detection removal metrics. Modeling building for practical application. Guidelines for developing Quality Metrics. Challenges in Instituting Metrics Programs. Benefits: Participants will learn practical and effective techniques for defining and using software metrics. They will be able to: distinguish “busy” metrics from truly useful metrics; use measurement to control current projects and plan for future projects; and assess the state of their organization’s existing practices. Practice with hands-on exercises, based on a case study and led by an instructor, will illustrate strategies for success. This course in Software Metrics complements all other courses in the SQI curriculum. It is particularly useful when taken with: Software Quality Assurance Inspection Techniques Managing the Testing Process Software Metrics also supports the Software Project Management Certificate Program courses: Foundations of Software Project Management Essentials of Software Engineering Principles of Software Testing Business Context and Legal Considerations Software Quality Assurance Management of Software Project Teams This course is especially useful for: Project managers, software engineers, software developers, technical leads, user representatives, software quality assurance professionals, testers, inspectors, and any stakeholder engaged in using measurement to improve the software development process and product.
Contact Information
Theresa
Dobbs
Sr. Manager
tdobbs@mail.utexas.edu
(512)
471-4875 |
Course Location
MCC Building, 1.108
3925 W. Braker Lane
Austin, TX 78758
Telephone: (512) 232-9440
For a map and information on
parking, visit this link. |
|
Attendance Policy
A 90% attendance rate is required in order to earn Continuing Education Units (CEUs) and certificates of completion.
|
Cancellation Policy
A full refund, less a $100.00 processing fee, will be given for all cancellations or transfers if notification is received on or before two weeks prior to the beginning of the course. Refunds will not be given for cancellations received after this date. Substitutions may be made anytime prior to the first day of class. The CLEE programs are subject to cancellation if too few people enroll, in which case a full refund will be given.
|
|
Hotel Information
|
| 
