Project Control (Part 3)

Lag and Leads

Lags and leads are techniques to introduce duration delays or accelerations in the network. These are used for all types of relationships, the logic of predecessor/successor is respected; however the start and end date calculations for each activity are adjusted according to the technique used (Forward and Backward Passes).

Lags will delay the successor’s start or end date (depending on the type of relationship, it has with its predecessor). Lags are often used when the successor activity, in order to commence, requires the predecessor activity to perform of its own work. At other times, the successor activity can only be completed after receiving and utilizing the results of its predecessor. Often lags are introduced in the network to cater for a lapse of time that does not use resources.

Leads will accelerate the start date of the successor activity. They are only used for FS links when seeking to optimize and fast track the project.

Figure 1.0 the application of lags and leads

Forward and Backward Passes

The activity network represents the logical flow of work to be performed. A completed and verified network illustrates the sequences of activities. These sequences are called path. All the paths radiate from the start point of the network and converge back into the end point.

Now it is time for the activity network to return meaningful scheduling dates for each activity to the team. The critical path method (CPM) provides for this. The method was developed in the 1950s in a joint venture between Du Pont Corporation and Remington Rand Corporation. While the scheduling method was developed specifically for the construction industry, it can be applied to any project with interdependent activities. CPM enables the project critical path to be determined by tracing the logical sequence of activities that directly affect the completion date of the project through a project network from start to finish. There may be more than one critical path depending on workflow logic. A delay to progress of any activity on the critical path will cause the overall project duration to be extended.

Using CPM, date calculations are made for each activity a long each path by proceeding by a Forward pass of the network and then a Backward pass (see Figure 2.0).

Figure 2.0

Each activity is then boxed by the following:

    1.     Earliest Start time – ES is the earliest time at which the activity can start depending on its precedent activities.

     2.     Earliest Finish time – EF is the earliest time at which the activity can finish. This is equal to the earliest start time for the activity plus the duration of the activity.

    3.     Latest Start time – LS is the latest time at which the activity can start. This is equal to the latest finish time for the activity minus the duration of the activity.

    4.     Latest Finish time – LF is the latest time at which the activity can be completed without delaying the project.

At this stage of scheduling it is important to note that all calculations will be made in absolute terms of duration in the established working unit, where the network’s start is set to zero.

The GANTT chart, once adopt to both civil and corporate calendars, will present the schedule against a meaningful calendar.

Forward Pass

The forward pass is the technique used to calculate the earliest start (ES) and earliest finish (EF) dates for each activity on each activity on each path of the network. It is a relatively simple process, requiring easy arithmetic (unless using PERT or other probabilistic estimating technique). The principle for single predecessor, finish to start (FS) relationships is straight forward:

The network start activity, of a duration of zero, is set to zero. Its ES and EF are also set to zero. Proceeding down each path, the ES of an activity is set to equal the EF of the predecessor activity. The EF of the said activity is set to its ES plus its duration.

This process is followed until all ES and EF dates of all activities are determined for all paths. As all paths converge to the finish activity, which has a duration of zero, the ES of the finish can be set and its EF set equal for the ES (see Figure 3.0).

Figure 3.0

Multiple Predecessor

When more than one predecessor exist, the ES of the activity will be set to highest EF value of its predecessor.

In the example (see Figure 4.0), the paint walls activity cannot start before both activities clean/dry walls an earliest finish (EF) date of 15 and paint delivery with an EF dates, the resulting ES date of paint walls become 36.

Figure 4.0

Backward pass

The backward pass is the technique to calculate the latest start (LS) and latest finish (LF) dates for each activity on each path of the network. It is a relatively simple process, requiring easy arithmetic.

The principle for single predecessor, finish to start relationships is straight forward:

The pass commences with the network’s finish activity date. The ES date of the finish activity has been calculated by the forward pass. Since the activity has a duration of zero, then its EF date is set equal to the ES. The LS and LF dates also set to equal the activity’s ES and EF.

Proceeding backwards along each path, the LF of an activity is set to equal the LS of the successor activity. The LS of the activity is set to its LF minus its duration.  This process is followed until all LS and LF dates of all activities are determined for all paths. As all paths converge back to the beginning of the network, the LF of the start activity can be set and the LS is set equal to it (see figure 5.0).

Figure 5.0

Calculating Float (Slack)

Float is the amount of time that an activity can be delayed past its ES or EF without causing a delay to:

          1.     Project completion date – Total Float (TF)

          2.     Subsequent activities – Free Float (FF)

Total Float

The Total float for an activity is the total amount of time that a schedule activity may be delayed from its EF date without delaying the project finish date, or violating a schedule constraint.

TF = LF – EF or LS - ES

Free Float

The Free float is the amount of time that an activity can be delayed without delaying the ES of any immediate successor activity.

The Free float is determined between two consecutive and is the difference between the successor’s ES start and the predecessor’s EF.