Project Management Project Management involves the planning, monitoring, and control of the people, process, and events that occur as software evolves from a preliminary concept to an operational implementation. The 4 P´s in Project Management: People, Product, Process, and Project

People People need to be organized to work effectively. The „people factor“ is very important. The Software Engineering Institute has developed the „People Management Maturity Model“ to promote organizations to attract, train, motivate and retain the talent needed for SW development. The model defines the tasks of recruiting, selecting, managing performance, training, compensation, career development, organization & work design, team/culture development. Curtis, B. et al., People Management Capability Maturity Model, Software Engineering Institute, 1994.

Product Scope and requirements must be understood, which means a good communcation with the customer must exist. (See Blastoff Document!) Once they are understood, alternative solutions are considered to find the best approach, given the constraints (time, budget, personnel, technical interfaces, etc.)

Process A suitable process must be selected because we need to manage complexity. The process provides a framework from which a plan for SW development can be established. Work tasks, project milestones, work products, quality assurance points, enable the activities to be adapted to the SW project and requirements of the team. Software quality assurance, SW configuration management and measurement are tasks that overlay the process model.

Project In 1998 industry data indicated: 26% of SW projects failed outright 46% experienced cost and schedute overrruns To avoid failure, a software manager & software engineers have to identify a set of common warning signs, understand the critical success factoirs for good project management and develop a commonsense approach for planning, monitoring and controlling the project.

Project (cont.) The project must be planned, estimating effort and calendar time to acomplish work tasks. Select the life-cycle model to be used Elaborate the different types of plans (configuration management, quality assurance, etc.) Establish mechanisms to monitor and control work. Document the plan (WBS, PERT/CPM) Use standards (DOD-STD-2167A, NASA-Sfw-DID- 02-ADA, IEEE Std ) Cost, duration of the project and resources have to be estimated.

Parameters for Planning Requirements CostTime At least one of these parameters has to stay under the control of the development team. Preferably two.

WBS (Work Breakdown Structure) (Planning Technique) Decompose a project into levels from complex to simple activities : Level 1: Name of the project/Objective Level 2: Subsections or departments Level 3: Tasks Level 4: Subtasks Install LotusNotes to support project development Install server Install Local client Install Remote client Design DB Program DB 1 2 3

Gantt Chart (Planning Technique) They show each activity and the starting and edn- points in time (estimations). 2. Analysis 3. Design 1. Planning 7. Testing Week 1Week 2Week 3Week 4Week 5Week 6Week 7Week 8.:.:

This methods help to define: Critical activities Critical path (series of critical activities) Starting dates (early or latest) Ending Date / Duration (aerly or latest) Important dates Analyze requirements 214 days 1/41/21 10 days 1/4 Revise Requirements 31 day 1/22 Starting meeting Risk Analysis 31 day 1/22 Initial cost estimation 31 day 1/ Critical Path Method (CPM) (Planning Technique)

CPM Parallel activities DiseñoGUI 614 days Tue 11/24/98Fri 12/11/98 DiseñoBasededatos 721 days Tue 11/24/98Tue 12/22/98 ProgramaciónBD 921 days Wed 12/23/98Wed 1/20/99 Pruebadeusabilidad 107 days Wed 12/23/98Thu 12/31/98 PruebadelaBD 117 days Thu 1/21/99Fri 1/29/99 Pruebadelsistema 1210 days Mon 2/1/99Fri 2/12/99 Escribirmanualde usuario 147 days Mon 12/14/98Tue 12/22/98 Entrenamientode usuarios 155 days Wed 12/23/98Tue 12/29/98 ProgramaciónGUI 87 days Mon 12/14/98Tue 12/22/98 Revisióndeldiseño 51 day Mon 11/23/98

SW Process and Project Metrics Quantitative measures that enable people to gain insight into the efficacy of the software process and the projects that are conducted. Basic quality and producticity data are collected and analyzed, compared and assessed. They help to pinpoint problem areas so that solutions can be developed, and the process can be improved. They have to be applied consistently, yet simple metrics that are never used to assess, reward or punish individual performance.

Product Metrics Size oriented metrics –Lines of code (LOC) –Errors per KLOC (thousand LOC) –Pages of documentation per KLOC,... Function-oriented metrics –Function Points (FP) –FP per month,... Logic Complexity Metrics –Cyclomatic Complexity –Knots,...