Software Project Management - PowerPoint PPT Presentation

1 / 46
About This Presentation
Title:

Software Project Management

Description:

Joint Project Planning (JPP) uses a goldfish-bowl approach to ... JPP is similar to Joint Application Design (or Joint Application ... for control freaks ... – PowerPoint PPT presentation

Number of Views:383
Avg rating:3.0/5.0
Slides: 47
Provided by: gle9
Category:

less

Transcript and Presenter's Notes

Title: Software Project Management


1
Software Project Management
  • Joint project planning controlling project
  • INFO 638
  • Glenn Booker

2
Joint Project Planning
  • Joint Project Planning (JPP) uses a goldfish-bowl
    approach to conducting early analysis of a
    project
  • Its scope is typically to define the POS and/or
    PDS
  • For software, this might include defining the
    system scope and key requirements, and/or
    developing high level system design

3
JPP vs. JAD
  • JPP is similar to Joint Application Design (or
    Joint Application Development) (JAD)
  • JPP is more general than JAD
  • JPP could be used for planning any kind of
    project
  • JAD is software-specific

4
Planning JPP
  • JPP needs to create an environment in which key
    decisions can be made about the project
  • Planning JPP is crucial to its success
  • It is critical that key people be required to
    attend a JPP session
  • Notice required, not invited
  • JPP doesnt work unless the players are all
    present

5
JPP Attendees
  • All significant stakeholders in a project need to
    attend JPP
  • The tough part is identifying what significant
    means
  • Attendees typically include
  • Facilitator an outsider whose role it to lead
    the JPP session
  • Typically have training in JPP, and are excellent
    negotiators

6
JPP Attendees
  • Project manager whoever will lead the project
    is an obvious choice to attend
  • Technographer a scribe who will record the
    results of the session
  • Might be proficient in tools for recording
    brainstorming sessions, prototyping a system, or
    other appropriate skills
  • Other key project people such as a system
    architect, managers who will report to the
    project manager, etc.

7
JPP Attendees
  • Customer rep is often included
  • Resource managers, such as IT staff, HR, or other
    relevant support personnel
  • Project champion (aka sponsor) whoever has been
    pushing to make the project happen, other than
    the manager
  • Process experts to help make sure the project
    will follow sound processes
  • Anyone else you deem necessary!

8
JPP Logistics Facilities
  • JPP needs to take place in an isolated
    environment, to help everyone focus on the same
    thing
  • Generally held offsite, such as a hotel or
    conference center
  • Typically allow a few days for the JPP, depending
    on the scope of the project and the goals of the
    session

9
JPP Equipment
  • JPP might use various tools to capture the
    results
  • MS Visio for process flowcharts
  • Axon or Mind Mapper for capture of brainstorming
  • These are in addition to the usual conference
    equipment computer projector, sticky notes,
    etc.

10
JPP Agenda
  • JPP needs to have a specific agenda defined
    before the session starts
  • The agenda must define what is expected to come
    out of the session
  • A completed POS, and/or
  • A completed PDS, and/or
  • A project plan, etc.

11
JPP Deliverables
  • More specific deliverables could include
  • WBS
  • Activity duration estimates
  • Resource requirements
  • Project network schedule (Pert)
  • Activity schedule (start/end dates)
  • Resource assignments

12
Project Proposal
  • The JPP might result in a project proposal,
    including
  • Background
  • Objective
  • Approach
  • Statement of work
  • Time cost summary
  • Appendices

13
Monitoring Progress
14
Monitoring Progress
  • By now you have been able to create a detailed
    project plan, including task definitions,
    estimates of duration, assignment and leveling
    of resources
  • Then the project actually starts
  • Now you need to monitor what really happens, and
    control the future of the project

15
An Aside
  • This is great stuff for control freaks
  • You get to define what people will do, when
    theyll do it, and even tell them who is their
    boss
  • Now you get to decide if they are doing their job
    right, and what youll do if they arent
  • Isnt this a great world?

16
Control and Risk
  • Controlling a project is closely linked to risk
    management
  • You want to minimize the risk that the project
    wont finish successfully
  • Successfully generally means on time and within
    budget
  • To do so, you need measurements to help decide if
    the project is on track

17
Use Pictures
  • Graphics are key to presenting information well
  • Most senior managers dont have time to read tons
    of words
  • A well thought out graphic will convey critical
    information quickly and with minimal explanation
  • If somethings wrong, need to address what
    corrective action will be taken

18
Controls
  • Without good controls, a project will wander like
    a 6-year-old on summer vacation
  • Controls allow us to
  • Track progress what has been accomplished?
  • Detect variance have we departed from the plan?
  • Take corrective action fix it!

19
Balance Control and Risk
  • More controls on a project
  • Costs more for planning and tracking
  • Reduces risks and creativity
  • So a critical question for every project is how
    many controls do I need?
  • Need enough to know whats going on, without
    micromanaging the project
  • The answer might change during the project

20
Balance Control and Risk
  • Too little control will increase project cost,
    because effort will be wasted
  • In theory theres an ideal balance possible
    between control and risk
  • Also need to consider that the product quality
    will also be affected by the amount of control
    over its development process

21
Progress Reporting System
  • Some form of progress reporting system needs to
    be established
  • Want timely, complete, clear, and accurate status
    reported
  • Avoid adding too much to overhead to create the
    status reports
  • Results are readily available
  • Warns of problems with time to fix them

22
Types of Status Reports
  • Typically there are five kinds of status reports
  • Current period reports report status during the
    current reporting period, e.g. this weeks status
  • Cumulative reports report history of project
    from start to the present, or at least a
    significant amount of time
  • Good for finding trends

23
Types of Status Reports
  • Exception reports are generated only when
    something is amiss
  • Summarizes whats wrong, and what action is
    desired to fix it
  • Stoplight reports arent really a separate kind
    of report
  • They add a simple red/yellow/green indicator of
    status green is all happy, yellow is a problem
    that needs fixing, and red means project is out
    of control

24
Types of Status Reports
  • Variance reports tell how far the project is
    ahead of, or behind the plan
  • Variances generally pertain to the schedule
    and/or costs, showing planned and actual values,
    and the difference between them
  • Present results from the current reporting
    period, and maybe one previous period
  • May be tabular data, or graphic

25
How When Collect Data?
  • Status reports are critical to understanding a
    project, yet can also be a waste of time and/or
    interfere with getting the project done
  • Need to decide how often reporting is done
  • Academia tends to be monthly, most of industry is
    weekly or biweekly

26
How When Collect Data?
  • Need to determine reporting period (what day is
    the start of the week?)
  • This often feeds a repeating process, e.g.
  • Reports are due Friday to your manager,
  • They report to their boss by Monday noon,
  • A collected report is issued on Tuesdays
  • Reports contain actual status to date, start and
    finish dates for tasks

27
How When Collect Data?
  • Reports might also include
  • Projections of work remaining,
  • Percent completion of tasks, and
  • The amount of resources spent on each task (e.g.
    12 hours on WBS task 1.3.2)

28
Variances
  • Variances are the difference between actual
    events and the project plan
  • Positive variances are often good
  • They mean you are ahead of schedule or under
    budget
  • But could mean the schedule has slipped, and
    little has been accomplished

29
Variances
  • Conversely, negative variances are generally bad
  • Behind schedule and/or over planned cost
  • Rarely, can mean more work has been done than
    planned

30
Displaying Status
  • There are three major ways to display the status
    of a project graphically
  • Gantt chart
  • Milestone trend chart
  • Cost schedule control chart

31
Gantt chart
  • We covered the Gantt chart in week 3
  • It is probably the most commonly used tool to
    plan and track projects
  • To show progress, dark thinner bars are used to
    show how much work has been accomplished
  • This example is 30 complete

32
Milestone Trend Chart
  • The Milestone trend chart is a plot of how well
    specific events and decisions (milestones) were
    accomplished
  • The horizontal lines represent 1-3 standard
    deviations above and below the expected
    completion date of each milestone
  • Presumably you have historic data to determine
    the standard deviations

33
Milestone Trend Chart
  • Like monitoring a control chart, poor trends
    (especially downward) or odd leaps in the data
    are keys to identifying problems

34
Milestone Trend Chart
35
Cost Schedule Control
  • Cost schedule control refers to the system used
    by the many agencies called earned value or
    C/SCSC
  • We have already defined a project plan with tasks
    and resources
  • Each task therefore has some defined dollar value
    (its resources times their hourly rate)

36
Cost Schedule Control
  • To use Cost Schedule Control, we need to define
    when we get credit for accomplishing each task
  • Only when the task is done
  • Half at the task start, and half at finish
  • Etc.
  • The total planned value of work done on the
    project typically forms a long S curve over time

37
Cost Schedule Control
  • The planned amount of work, in terms of its
    value, over time form a curve called Planned
    Value (PV) (formerly BCWS)
  • As the project happens, we record the actual
    costs incurred (AC) and how much work we really
    got done (EV) (formerly ACWP and BCWP)

38
Cost Schedule Control
  • We can find variances in terms of cost (related
    to whether we finish within budget) and schedule
    (will we finish on time)
  • At any time during the project
  • Cost variance AC - PV
  • Schedule variance EV PV
  • Recall that negative variances are bad

39
Cost Schedule Control
  • We can also define indices to tell us if we have
    a good trend in getting work done
  • Schedule performance index SPI EV/PV
  • Cost performance index CPI EV/AC
  • Indices gt1 are good (got more work done than
    planned or budgeted)

40
Cost Schedule Control
  • So monitoring a project with cost schedule
    control generally means using
  • A plot of PV, AC, and EV versus time
  • Plots of cost and schedule variances
  • Cost and schedule performance indices
  • Based on these, look for negative variances
    and/or indices lt one

41
Status Detail
  • The amount of detail in status reporting depends
    on the management level of its audience
  • First line managers generally want lots of
    detail
  • Project managers generally want to focus on
    problems they must resolve
  • Senior managers need a very brief synopsis of
    status

42
Status Meetings
  • Some form of meeting is often desired to
  • Share the current status of each part of the
    project
  • Look for collaboration opportunities
  • Make decisions about problems

43
Meeting Minutes
  • Record the actions and decisions in a meeting
    with minutes
  • Identify who was there
  • Identify what happened
  • Review previous action requests
  • Review old issues
  • Review new issues
  • Identify what new actions need to be followed up
    on, by whom, and when

44
Change Control
  • A change control process is needed to manage
    changes to the scope of a project
  • See this example from the FAA
  • The example cited was used for managing problems
    reported with an air traffic control system, but
    the basic principles are universal

45
Change Control
  • It describes the activities needed to analyze a
    problem, estimate how much work it is to resolve,
    determine its priority, fix it, and incorporate
    it into a system change with other problem fixes
  • The names of the organizations which perform each
    of the steps may vary wildly, but some sort of
    review board or change control board is typically
    used

46
Escalation
  • Escalation here means how a problem can be
    resolved
  • Little problems might be resolved by the project
    manager
  • Bigger problems might be resolved by getting
    additional resources from your organization
  • Huge problems might need cooperation from your
    customer to resolve
Write a Comment
User Comments (0)
About PowerShow.com