Civil Engineering Reference
In-Depth Information
The scheduling process is iterative. Each iteration cycle either adds or relaxes con-
straints, but the iterations should always be guided by the goals and constraints identi-
fied in the master strategic planning document. It is not inconceivable, however, to need
replanning after unsuccessful attempts to produce a viable schedule. Hence, there is an
inherent feedback mechanism between the planning and scheduling processes.
There are many users of the information produced by the scheduling process: the
owner, the design-builder, the engineer, the subcontractors, the suppliers, and even the
public. Commercial scheduling software can offer multiple ways of packaging, display-
ing, and communicating scheduling information to accommodate different needs of many
consumers. Bar charts, time-scaled logic diagrams, network diagrams, project strips, 4D
animations, and tabular reports are all common means of conveying schedule information.
Basic uses of scheduling information include predicting project completion, deter-
mining when tasks can and must start and finish, predicting resource demands as well as
mitigating supply/demand conflicts, detecting conflict among self-performed trades, and
computing progress payments. Beyond the basic uses of scheduling information, well-
prepared schedules can and should also be used to tightly coordinate the work of subcon-
tractors, multiple primes, and owner-supplied information. Creating an as-built schedule
and evaluating the time impact of changes are two advanced applications of scheduling.
In scheduling, there are two ways to represent tasks in a network, namely by nodes
and by arrows. The former is very intuitive, hence preferred, and it is known as the pre-
cedence diagramming method (PDM). The latter is known as the arrow diagramming
method (ADM) and requires the use of logical “dummy” and naming “dummy” tasks,
which makes it less intuitive. Although ADM was the introduced first (in the 1950s), it
has been gradually superseded by the PDM (since its introduction in the 1960s). Unfortu-
nately, it is now difficult to find commercial software that supports ADM.
Notwithstanding the network notation, the critical path method (CPM) computes the
critical path (i.e., the longest path or paths in the network), identifies the tasks belonging
to such path(s), and identifies the noncritical tasks. A direct product of the CPM is the cal-
culation of total float (TF), which is the time difference between when a task can start/fin-
ish and when it must start/finish to avoid delaying the project completion time. Whereas
tasks that are in the critical path have by definition no TF, noncritical tasks do have posi-
tive TF. A key property of TF is that it is shared by the tasks belonging to the same non-
critical path. This is of great significance, because if upstream path tasks consume some
or all of the TF, this used-up TF is no longer available for subsequent downstream path
tasks. Because TF is shared, there exists a “first come, first served” functionality. When
there are multiple parties associated with this chain of noncritical tasks, consumption of
TF by one party upstream is at the expense of the others downstream.
A type of float that is not shared but belongs to each noncritical task is known as
free float (FF). Free float is defined as the amount of time any activity can be delayed
without impacting the early start time of the task's immediate successors. A task's FF is
less than or equal to the task's TF; consumption of FF reduces the amount of TF by the
same amount.
Plain Vanilla Schedules™ are characterized by continuous tasks with finish-to-
start (FS) relations with no lags, one calendar, no resources, and no start/finish exter-
nal constraints. The modeling capabilities of these simpler schedules can be enhanced
by the addition of start-to-start (SS) and finish-to-finish (FF) relationships with lags, by
Search WWH ::
Custom Search