Project Planning

CITS3220 Software Requirements &

Project Management
Why plan ?
! The superior programmer uses
superior planning skills
to avoid having to use
superior crisis management skills
! British Army Saying (The 6 P's)
Proper Planning Prevents Piss Poor
Performance
Issues in Project Management
(a reminder)
! Before starting the project:
" planning, estimation, risk analysis;
! While doing the project:
" monitoring, evaluation, metrics;
! After the project is done:
" assessment and improvement
Overview of Todays Lecture
! Issues
" Why plan? Whats in a plan? Human factors
! Definitions and Techniques
" planning steps; work breakdown structure;
critical path analysis and PERT charts; Gantt
charts
Why Plan ?
! Control of the project is based on derivation from the
plan
! Communication of roles, responsibilities and resource
requirements
! Strategy enables trade-offs in the cool light of day rather
than within a deadline panic
! Risk management anticipates problems and provides a
management strategy
! Granularity how much planning to do?
A plan is a living document
Questions for Project Managers
! What is the scope of our project?
! What gets delivered for the available time,
people, $? (And what won't we do?)
! What resources do we need?
! How do we tell where we are?
! Are we on schedule?
! Are we on budget?
! How do we communicate our plan inside &
outside the team?
The W
6
H
3
elements of a plan
! Objectives
" WHY is the system being
built?
! Deliverables
" WHAT has to be produced?
! Milestones
" WHICH process stages are
needed for control?
! Methods
" HOW will the tasks be done
(technically)?
! Resources
" WHERE do we get the
people/tools for the work?
! Risks
" HOW will potential problems
be minimised?
! Schedule
" WHEN will the stages be
completed?
! Verification
" HOW will we know the
system is correct?
! Responsibility
" WHO is responsible for
ensuring completion?
An effective plan includes
! Business perspective financial, technical commercial
! Deliverable components and the tasks required
! Clearly defined organizational responsibilities
! Product requirements and deliverables
! Risk quantified in $ or hours and risk management
strategy proposed
! Sensible task durations e.g. 2 months max
! Sensible task size (e.g. 1 to 2 person-months)
! Defined acceptance criteria
A poor plan
! only technical perspective or only task view
! no clear assignment of responsibility for
production of deliverables
! no milestones or acceptance criteria
! risk ignored or buried in bloated estimates
! long task durations (e.g. 1 year)
! very large tasks (eg 20 people for 2 months)
! no process to guarantee the clients needs are
met
Human Factors in Project
Planning
! people tend to be risk averse when there is a potential of
loss
! people are unduly optimistic in their plans and forecasts
! people prefer to use intuitive judgment rather than
(quantitative) models
! Managers are people also, and these human tendencies
apply to the managers who make business decisions at
the project to the executive level.
Tasks, Milestones &
Deliverables
! Tasks
" activities which must be completed to achieve
the project goal
! Milestones
" important checkpoints
! Deliverables
" milestones with an external output
Work Breakdown Structure
(WBS)
! A to-do list, sorted by category
! Task description (what)
! Estimated time (length)
! Person responsible for task (who)
! Resources required
! Cost ($)
3 Steps to a Project Plan
! Step 1: Identify Tasks & Resources
! Step 2: Identify Milestones
! Step 3: Analyse Dependencies and
Define Schedule with a Gantt Chart
Initial Plan, Tasks and
Resources
! Tedious and hard......but must be done
! What ==> tasks
! When ==> schedule
! How ==> people, materials, equipment
costs
! Set scope and avoid "scope creep"
Identifying Tasks & Resources
! State each task using "verb-noun" form
" Examples: Write manual; Implement Prototype;
Review requirements with client
! Use appropriate level of detail
! Function, not form, known at start of project
" Example: "Build concept demonstration prototype"
! Make each task significant
" e.g. Identify competitive products" not "Go to library
Defining tasks is hard but worthwhile!
Milestones
! Key targets for completion of certain phases of a project
! A milestone requires both the state of the task/phase
and a date/time to be set.
! Some typical SE milestones
" requirement specifications completed
" key plans completed
" systems documentation completed
" test data sets completed
" tests run with user prepared test data completed
" user manuals completed
" acceptance testing completed
" post-implementation review completed
Milestones
! Types of milestones
" Provide tangible interim goals
" Demonstrate progress and so provide
motivation
" Used to enforce schedule
! State each milestone in "noun-verb" form
Examples: Mission stated; Mid-quarter
review presented; Prototype completed
Scheduling &GANTT charts
! Gantt charts
" Best basic scheduling tool
" Developed in 1917 by Henry L. Gantt
! Pick appropriate time scale (days/wks)
! Method of tracking progress
! Who is responsible
! Good software (e.g. MS Project) is available.
! GANTT charts dont show dependencies
explicitly
Source http://searchcio.techtarget.com/
Gantt Chart components
! horizontal axis = total time span of the project
! vertical axis = project tasks
! horizontal bars = sequences, timing, and time
span for each task
! bar spans (tasks) may overlap
! secondary bars, arrowheads, or darkened bars
added as project progresses to indicate
completed tasks
! vertical line = report date
PERT charts
! Programme Evaluation & Review
Technique
" Explicitly shows dependency and parallel
tasks
" Shows critical path and where flex time is
" Can be hierarchical (with sub-tasks)
Source http://searchsmb.techtarget.com/