Scheduling

1
Scheduling
Problems with poor scheduling Scheduling
techniques Negotiation tactics
2
Scheduling and Estimation

• Scheduling uses estimates of effort for a project
  to allocate and prioritise tasks and people
  effectively
• if the estimates are poor, this effects the
  scheduling
• over estimating effort leads inflated project
  costs
• underestimating leads to insufficient time and
  resources being allocated

3
Over optimistic schedules

• Under estimating effort can lead to deadlines
  that are too tight, in order words the schedule
  will be too optimistic
• limited time and resources will effect quality
• there are other factors which may lead to over
  optimistic schedules

4
Causes of poor scheduling

• Setting schedule before defining requirements
• refusal to accept an estimate
• aggressive scheduling to win bids or save on
  resources
• target setting
• adding requirements
• cannot/will not change deadline

5
Effects of over optimistic schedules
Impacts on

• schedule accuracy
• ability to stick to plan
• quality of plan
• scope of project
• quality of project
• customer relations

6
Effects on quality of plan

• Impact can be large and long lasting and even
  affect other projects
• Prevents effective planning by feeding wrong
  information into
• phase planning
• resource allocation (staffing included)
• development and testing planning

7
Quality

• May end up missing out on some fine details or
  even functionality
• may rush requirements and analysis or testing
• spend time on surface details to look more
  polished (e.g. user manuals, interface)
• release before product is ready

8
Quality

• May attempts shortcuts in development to meet
  deadline
• ignore maintainability
• remove or turn off features
• fix only most obvious defects
• skimp on documentation which will have impact in
  future
• testers find defects faster than developers can
  correct them
• no time to carefully implement fixes, causing
  more errors

9
Customer relations

• Will blame those working on project
• Need for reassurance may be a distraction
• May look at project with a more critical eye
• Penalty clauses
• May be unhappy with quality

10
Effects of pressure to meet deadline

• Compromises on quality could lead to project
  taking longer than it should
• hard for staff to work to optimum productivity
  and quality if stressed and overworked
• staff turnover may increase
• adding more staff may lead to decrease in
  productivity

11
summary of disadvantages of poor scheduling

• Undermines effective planning
• creates unnecessary pressure
• Affects developer morale
• Affects customer relations
• Compromises quality
• creates situation where promises cannot be
  delivered
• likely to result in longer schedule

12
Better schedules

• Improved estimation
• Improved scheduling techniques
• Improved negotiation

13
Improved estimation

• See previous lectures
• combination of expert judgement and estimation
  techniques seem to be most promising
• local models using local data make better
  estimates

14
Improved scheduling

• Will use estimates of time/effort, so very much
  dependent on quality of estimates
• Use PERT/CPM to show
• how where activities can be carried out
  concurrently
• critical path
• slack
• Gantt charts display much of same information, so
  useful to cross check

15
Scheduling Techniques

• WBS
• PERT
• CPM
• Gantt

16
Work Breakdown Structure

• Development of project plan is predicated on
• having a clear, detailed understanding of the
  tasks involved
• having an estimate of how long each will take
• knowing the order in which the tasks should be
  performed
• knowing the dependencies between each task
• knowing what resources are available so each task
  can be assigned to the appropriate person

17
Work Breakdown Structure

• A hierarchical decomposition of a project into
  lower levels
• each item at a specific level is numbered
  consecutively and each lower level decomposition
  numbered within the higher level item (e.g 10
  10.1, 10.2, 10.3)
• broken down until a manageable size task is
  arrived

18
Program Evaluation and Review Technique

• Charts showing task duration and dependencies
• can show milestones, checkpoints and reviews
• small projects may be shown on one charts, but
  larger ones will be broken up into more
  meaningful parts
• probabilistic

19
Critical path method

• Same sort of chart as PERT, with critical path
  highlighted. On a superficial level, they can be
  considered the same
• Deterministic not probabilistic
• critical path is the series of connected
  activities which takes the longest time in total
• can change during the course of the project, if
  changes are made, tasks fall behind or finish
  ahead of schedule

20
Gantt chart

• Matrix listing tasks to be performed on vertical
  axis and duration shown on horizontal axis
• tasks represented by a bar may labelled with or
  colour coded to indicate which individual is
  responsible or may have separate section to show
  personnel allocations

21
Drawing a PERT chart

• Nodes (circles) represent the beginning or end
  of activities
• Must have a start and end node for each chart
• Arcs (arrows) represent the tasks (activities)
  and are labelled with the name of the the task
  and its duration
• Path is a sequence of connected activities from
  start to end

22
PERT Chart
Node (event)
start node
end node
activity name
A
B
START
END
3
4
activity
duration

• Notations will vary slightly, but will contain
  the same basic elements
• If no start and end nodes are shown we will
  assume the example is part of a chart
• NB Some show activity on node

23
Dependencies
x
z
y
Activity z cannot begin until activities x and y
have been completed
24
Activity List
Activity Description Dependent on Time A
insert description here 2 weeks B 3
weeks C A 2 weeks D B 4
weeks E C 4 weeks F C 2 weeks G D,
E 4 weeks H F, G 2 weeks
25
Example chart
What is the critical path?
START
END
26
Critical path highlighted
2244214
C
1
3
2
F
A
2
2
H
0
6
E
4
END
5
2
3
4
B
G
4
2
4
D
27
Building PERT Project Network
START
How will we add activity K, estimated time 9,
which is dependent on E and J
28
Dummy Activity
H
F
5
7
9
5
8
4
E
G
I
5
7
A
B
C
0
4
1
2
10
2
4
10
D
6
Adding a dummy activity shows how one activity
(K) must follow another (E)
7
9
6
8
11
J
K
29
Forward Pass

• Earliest start is 0 for activities with no
  predecessor
• Ej time event j will occur if the preceding
  activities are started early as possible
• Forward pass compute Ej for each event j
  beginning at the first node and moving forward,
  taking into account dependencies
• Earliest time rule - Ej is the maximum of the
  sums Ei tij involving each immediately
  preceding event i and intervening activity ij

30
Earliest Time
31
Example
2
4
C
1
3
2
F
A
0
12
14
2
2
H
6
E
4
5
0
2
END
3
4
START
B
G
4
2
4
D
3
8
32
Backward pass

• LF (latest finish) is EF for activities with no
  successor
• Li latest time event i can occur without
  delaying completion of the project
• Backward pass compute Li for each event i
  beginning at the last node and moving backward,
  taking into account dependencies
• Latest time rule - Li is the minimum of the
  differences Lj tij involving each immediately
  following event j and intervening activity ij

33
Latest time
1
L1
ti1
L2
ti2
i1
j
2
i3
i2
Lj
3
ti3
L3
34
Example
2
4
C
1
3
2
F
A
0
12
14
2
2
H
6
E
4
5
0
2
END
START
3
4
B
G
4
2
4
D
3
8
35
slack
Activities not on the critical path do not have
to start on the earliest date - there is some
slack
slack time you can delay activity without
violating latest finish ie LF-EF ES earliest
start time for a given activity length of
longest path from START to tail of activity
arc LS latest time a given activity can start
without delaying the project Slack time for
activity ij LS - ES(Lj - tij) - Ei
tij
j
i
Ej , Lj
ij
Ei , Li
36
Problems with CPM

• CPM is deterministic
• cannot make probabilistic estimates for project
  completion e.g we are 90 certain that it will
  be completed by this date.
• using the most likely estimate for activity
  completion but ignoring the variance, means it
  likely our CP estimate will be too short. (i.e.
  something always will go wrong)
• Conservatism often tends to create overlong
  projections

37
PERT Method

• Identify activities, estimates, and
  dependencies.
• Calculate the expected time and variance for
  each activity, using optimistic, most likely and
  pessimistic estimates (if making probabilistic
  estimate)
• Draw the network diagram
• Do Forward and Backward Pass
• Establish slack and the critical path
• Calculate project completion statistics
• expected duration is sum of expected activity
  times on critical path
• variance of duration is sum of variances on
  critical path
• Commonly probabilistic aspects are ignored,
  making this synonymous with CPM

38
Probabilistic estimates
use 3 time estimates to reflect uncertainty a
optimistic time (chance of finishing earlier) m
most likely time (best guess or mode) b
pessimistic time (chance of finishing later)
39
Formulae
Most likely time
? 2
variance
40
Probabilistic Time Estimates
Area under curve between a and b is 99.74
41
Example
activity duration predecessor expected
duration A requirements analysis 2/3/6
weeks - 3.33 B programming 3/6/10
weeks A 6.17 C get hardware 1/1/2
week A 1.17 D train users 3/3/3 weeks B,
C 3.00 CRITICAL PATH A-B-D EXPECTED
DURATION 3.336.17312.5 VARIANCE (6-2)/62
(10-3)/62(3-3)/621.805 STD 1.344
42
Crashing
Crashing reduces project time by expending
additional resourcesCrash time is the amount of
time an activity is reducedCrash cost is the
cost of reducing an activitys timeThe goal of
crashing is to reduce a projects duration at
minimum cost
43
Gantt charts
NB Gantt is a name not an acronym, so no need to
capitalize
44
(No Transcript)
45
Summary
Scheduling Techniques such as PERT, CPM and Gantt
charts are intended to assist the project manager
with organizing and resourcing the project Based
on largely the same information, displayed in
different ways Graphical representation - helps
PM visualize how activities interact Will only be
as good as the estimates made
46
Negotiation

• Project manager needs to negotiate realistic
  schedule and resources
• PM may be asked for estimates, but often have to
  negotiate with higher management, marketing or
  customer
• could use negotiation techniques e.g. principled
  negotiation

47
Principled negotiation
 Conflict resolution technique
(win-win)  Avoids positional bargaining and
assumes no dispute is win-lose (not all agree)
Fisher, R Ury, W. Getting to Yes Negotiating
Agreement Without Giving In 1981
48
Positional Bargaining vs principled negotiation
positional bargaining
principled negotiation

• Participants are adversaries
• victory over opponents is the goal
• demand concessions as condition of relationship
• hard on people as well as problem

• Participants are problem solvers
• goal is acceptable outcome for all
• separate people from problem
• hard on problem, soft on people

49
Positional Bargaining vs principled negotiation
positional bargaining
principled negotiation

• Distrust others
• entrenched position
• make threats
• demand one sided gains as price of agreement

• Proceed independently of trust
• focus in interests not positions
• explore interests
• invent options for mutual gain

50
Positional Bargaining vs principled negotiation
principled negotiation
positional bargaining

• Search for single answer - the one that suits you
• insist on your position
• try to win battle of wills
• apply pressure

• Develop multiple options to choose from
• insist on using objective criteria
• emphasise objective criteria over will
• yield to principle not pressure

51
Four themes of principled negotiation

• separate the people from the problem
• focus on interests not positions
• invent options for mutual gain
• insist on objective criteria

52
Separate the people from the problem

• Separate relationship issues/people problems from
  substantive issues, and deal with them separately
• people problems include
• issues of perception
• differing interpretations of reality
• emotions
• communication problems
• not talking
• not listening
• misunderstandings

53
Focus on interests not positions

• What people really want and need vs. what they
  say they want and need
• People may take extreme positions to counter
  their opponents positions

54
Invent options for mutual gain

• Follows from focussing on interests
• Find solutions to allow both sides to win
• Brainstorming
• no judgements
• record all ideas
• After brainstorming
• note most promising ideas
• evaluate according to mutual interests/needs,
  objective criteria

55
Objective criteria for decisions

• Can simplify negotiation process
• look outside for guidance
• do this step before evaluating possible solutions

56
Summary
Scheduling involves Estimating Planning Decision
Making Monitoring Control Negotiation Risk
management
It brings together most of the project management
techniques we have covered