Monday, January 11, 2016

PMP Prep: Resource Leveling and Resource Smoothing



Starting with its Guide to the Project Management Body of Knowledge (PMBOK®) 5th Edition, PMI® introduced resource smoothing as a resource optimization technique in the time management knowledge area. In earlier additions of PMBOK only resource leveling was available. A number of misconceptions exist with respect to resource leveling and resource smoothing:
  • Resource smoothing and resource leveling are similar terms and can be interchangeably used. According to the PMBOK, that's not correct. If they were same, the guide would have put them as synonymous words.
  • Resource smoothing happens after resource leveling--also incorrect. Resource smoothing doesn't have to happen after resource leveling.
  • Resource smoothing gives extra breathing space to project managers while scheduling. In fact, it's the opposite! Resource smoothing lessens the flexibility of the schedule.
These misunderstandings can create problems for test takers.

To understand resource leveling and resource smoothing, you need to understand two kinds of scheduling: 1) resource-limited or resource-constrained scheduling; and 2) time-constraining scheduling.

Resource Constrained Scheduling (RCS)
In RCS, a project is meant to be completed with the available resources, hence the name, "resource constrained." There will be an increase in the duration of the project if resources aren't available.

RCS considers both the supply (availability) and demand (requirement) of the resources. If the resource requirement exceeds the resource availability at any time during the schedule network analysis, some of the activities may be delayed until there's enough resource availability. This in turn may increase the duration of the project.

Note: Resource leveling is used in resource constrained scheduling.

Time Constrained Scheduling (TCS)
TCS, on the other hand, emphasizes the completion of a project within a specified amount of time. The time is determined during schedule network analysis. In TCS, project start and end dates are important and need to be respected.

TCS also considers both the supply (availability) and demand (requirement) of the resources. Here, however, there is a predefined limit on demand of resources, which can't be exceeded.

Note: Resource smoothing is used in time constrained scheduling.

Resource Leveling
Now, let's examine the definitions of resource leveling and resource smoothing according to the PMBOK. Resource leveling is defined as, "A technique in which start and finish dates are adjusted based on resource constraints with the goal of balancing demand of the resources with available supply."

As the guide explains, resource leveling can be used when:
  • Shared or critical resources are in limited quantities;
  • Shared or critical resources are available at certain times; or
  • Resources are over-allocated.
In resource leveling, the duration of the project can be changed. If resources are assigned to the activities on the critical path and are in limited quantities, it's likely that the critical path will increase in duration as we try to balance the demand for the resources with that of the supply. A similar situation can arise when the resources are available at certain times. Also, a project manager typically doesn't look for situations where resources are over-allocated or under-allocated. Rather, resources should be properly allocated and should work to their optimum capacities. If over-allocation (also called overloading) is there, then resource leveling is performed. 

Example: 
Using Microsoft Project, I've created the following bar diagram during network analysis. The human resources (R1, R2, R3, etc.) needed for the project are shown next to the bars in the diagram


As you can see, resource R1 is needed by activities A and C in the first three weeks of the project. That means R1 is over-allocated. The critical path for the project is "Start | C | D | Finish," which has been highlighted in red.

Let's apply resource leveling so that all resources are properly allocated. Post-leveling, the chart changes to the following:


The length of the original critical path has increased from seven weeks to ten weeks.

Resource Smoothing
The PMBOK defines resource smoothing as, "A technique that adjusts the activities of a schedule model such that the requirements for resources on the project do not exceed certain predefined limits."

It means we want to have a constant resource usage (resource profile) over time. The reasons are obvious. When there are high fluctuations in demand of the resources during a project, project cost may increase because you may have to hire them to cover the peaks in the resource profile. Also, when there are valleys (or troughs) in the resource profile, resources will remain idle during those periods while still being paid. Both situations are undesirable. Hence smoothing is needed and applied. The name "smoothing" comes from the fact that the peaks and the valleys in the resource usage profile are smoothed out.

In resource smoothing, the requirement of the resources don't exceed certain predefined limits. The critical path won't be touched to ensure that the duration remains unchanged. It means we can adjust the resource profiles within the available float (or slack) for the given activities. If the floats are in short supply for the activities, then we may not be able to optimize all the resources.

Example: 
Look at the simple network diagram shown below. Duration for the activities represents weeks.


Using forward and backward pass calculation, the critical path runs this course: "Start | C | D | E | F | Finish." All activities on the critical path will have float (or slack) of value zero. For activity B, the total float is 11 weeks; you could delay activity B by 11 weeks without delaying the project. Similarly, for activity H the total float is 12 weeks; you could delay activity H by 12 weeks without pushing the end date of the project.

In the diagram below, the required number of human resources has been shown along with the activity names in the diagram. "B(2R)" means activity B has a requirement of two resources.


At this stage, let us check the resource usage profile over the weeks. The number of resources need in each are added up (also called resource aggregation) and shown below the x axis under weeks. When put into a spreadsheet, the profile usage looks like this.


There are peaks in weeks 4, 5 and 6 and valleys in the latter part of the project. As I outlined earlier, these aren't desirable situations. Consider we that have a predefined resource limit, which is no more than seven resources in a week. Let's apply resource smoothing.

I can start activity B later since it can be delayed by 11 weeks without any impact to the project schedule. Similarly, activity H also can be started towards the end without changing the critical path. After making the changes, we get the following diagram.


When you lay that out in spreadsheet form, you can see that the valleys and peaks of resource usage have been smoothed out. It must be noted that in some situations, you may not be able to fully optimize the resources.


You should also note that after resource smoothing, the critical path has remained unchanged. However, the total float (and also free float) for activities B and H have been reduced. Hence, the flexibility for scheduling is now less for the project manager.

To help you remember the similarities and differences between the two resource optimization techniques, I've outlined them in the table below.

Similarities:

Differences:

This article was first published by MPUG on 3rd November, 2015.


No comments:

Post a Comment

Any comment is welcome - comments, review or criticism. But off-topic, abusive, defamatory comments will be moderated or may be removed.