Structure of Spm
Structure of Spm
Organizing functions deals with devising roles for individuals & assigning responsibility
for accomplishing project goals. Organization is motivated by need for co-operation when the
goals are not achievable by a single human being in a reasonable amount of time. The task
of organizing can be viewed as building a team given
A group of people
C
A common goal
What is best role for each individual ?
How responsibilities should be divided ?
The main issues affecting the project organization are:
The nature of task & how much communication the different team members need
to have among themselves.
Size of the team
LIBRARIAN
SPECIALISTS
PROGRAMMERS
The figure shows a graphical representation of the patterns of control & communication
supported by this kind of organization. The need for programmers & consultants as well as
what task they perform is determined by the chief programmer who initiates and controls all
decisions. The software library maintained by the librarian is the central repository for all the
software, documentation & decisions made by the team.
This team has a single point failure as all comunication must go through & all
decisions must be made by the chief progranmer. The chief programmer may become
overloaded or saturated. The success of this team organization depends clearly upon the skill
& ability of the chief progran1mer & the size and complexity of the problem.
3.8 SOFTWARE ENGINEERING
3.3.2 DECENTRALIZED CONTROL TEAM ORGANIZATION
In this team structure, decisions are made by consensus and all work is considered
group work. Team members review each others' work and are responsible as a group what
other members produce. Such a team organization leads to higher morale and job satisfaction
and hence to less turnover. The engineers feel ownership of the project and responsibility for
the problem, leading to higher quality in the work. Adecentralized control organization is more
suited to long ten projects because the amount of intragroup comnunication that it encourages
leads to a longer development time. It is suitable for less understood &more complicated
problems because a group can invent better solution than a single individual. It is based on
a tecunique called egoless programming as it encourages programers to review &share
each others' work.
The figure shows a graphical representation of the patterns of control & communication
supported by this kind of organization. The ring like management structure is intended to
show the lack of a hierarchy and that all team members are at the same level.
The negative side of this structure is that it is not appropriate for large teams, where
the communication overhead can overwhelm all the engineers, reducing overall productivity.
It also runs the risk of establishing a group that is forever in futile search of a perfect solution
to please everyone.
3.3.3 MIXED CONTROL TEAM ORGANIZATION
Mixed control team organization attepts to combine the benefits of centralized &
decentralized control, while ninimizing or avoiding their disadvantages. This organization
distinguishes the engineers into senior & junior engineers. Each senior engineer leads a group
of junior engineers. The senior engineer in turn rep orts to project manager. Control is vested
in the project nanager & senior engineers while communication is decentralized among each
st of individuals & their immediate supervisor. This mode tries to limit the comumunication
tu within a smallgroup. It also tries to realize the benefits of group decision making by vesting
authority in a group of senior programmers.
SOFTWARE PROJECT MANAGEMENT 3.9
shown
The patterns of control & communication in the mixed control organization are
in the above diagrams.
3.4 PROJECT CONTROL
START Design
Coding
Testing
The above figure shows the time required to complete different activities of a software
deveiopment project in the form of Gantt Chart. A Gantt chart helps in scheduling activities
but it does not help in identifying them.
3.4.3 PERT (PROGRAM EVALUATION & REVIEW TECHNIQUE) CHARTS
This chart is a network of bOxes or circles & arrows. Each box represents an activity
& the arrows are used to show the dependencies of activities on one another. The activity
at the head of an arrow cannot start until the activity at the tail of the arrow is finished. The
activities in this chart can be labelled with starting & ending dates of the activity. Some boxes
can be designated as milestones. A milestone is an activity whose completion signals an
important accomplishment in the life of a project. On the other hand, failure to make a
milestone signals trouble to the manager and requires an action to deal with the deviation
from the schedule.
To build a PERT chart for a project, one must first list all activities
required for
completion of the project and estimates how long cach will take. Then one must determine
the dependence of the activities on each other. The PERT chart is a
of this information.
graphical representati0n
Jan 1,98 Jan 3,98 Mar 10,98 Aug 5,98
The above figurc shows the activitics of a softwarc devclopment project , their starting
& cnding datcs, their interdependencics in the form of PERT Chart.
The advantages of using PERT chart
It forces the manager to plan
lt shows the interrelationship among the tasks in the projcct
It shows the critical path of the projcct thus hclping to focus on it.
It exposes all possible parallelism in the activities and helps in allocating resources.
It helps in scheduling
It enables the manager to nonitor &control the project.
The disadvantages of PERT chart are given below:
The use of this techniquc does not automatically guarantee the success of the
project.
It is not suitable for big projects as a PERT chart for such project may contain
hundreds of nodes.
Revision of such charts is necessary. So computer support is essential.