Critical Chain Solar Project Management
Inside this Article
As margins in the solar industry have decreased to single digits, contractors are looking for ways to control soft costs. Two crucial ways are to ensure that you do not have to return to a site to fix a mistake, and to make sure that you are not holding up the next task in line. This is where tight project management controls come in. In this article, I investigate a project management style called critical chain project management, or critical chain, which can help your team hit project deliverables on time and on budget.
First I review the progression of project management styles that informed the development of critical chain. Then I explain the mechanics and philosophy of critical chain. I conclude with a brief example of how to deploy it. Results will be best with complex and long-duration projects such as commercial and utility-scale projects. However, residential solar contractors can learn a lot from critical chain that they can apply to their projects.
A Brief History of Project Management
Critical chain project management is the culmination of improvements in project management methodologies dating back to the early 20th century. These earlier methodologies include the Gantt chart, the program evaluation review technique (PERT) and the critical path method.
Gantt chart. The most common and simplest type of project management is the Gantt chart. Developed by Henry Gantt circa 1910 and used extensively in World War I, a Gantt chart is an illustration of a project schedule using a bar chart to show the start and finish dates of a project. Project managers start by developing a work breakdown structure consisting of multiple work packages, as shown in Figure 1. Today, most Gantt charts also show relationships of dependency between various activities; some charts also have features, such as percent-complete shading bars, that allow users to view the current project status. (For more information, see “Managing PV Installations with a Gantt Chart,” SolarPro, October/November 2013.)
PERT. In 1957, the US Navy Special Projects Office developed PERT to support the large and complex Polaris nuclear submarine project. PERT is a method of analyzing project tasks and completion times to identify the minimum time needed to complete the total project. Because the technique accommodates uncertainty, users do not need to have a precise understanding of project details and activity durations to schedule a project. Instead, they employ best case, worst case and average timelines for tasks. Unlike a Gantt chart, which is oriented around start date, completion date and percentage of completion for individual tasks, PERT charts focus more on displaying the dependencies between tasks, as shown in Figure 2.
Critical path method. Morgan R. Walker and James E. Kelley of Remington Rand developed the critical path method in the late 1950s. Building on the project network framework developed for PERT, critical path is the longest progression of tasks dependent upon each other within a network of all project tasks. This project management methodology is premised on documenting the actual process of completing a project rather than a theoretical process. It documents all the project activities, the duration of each activity, dependencies between activities, and the major milestones and endpoints.
Critical chain method. Developed in 1997 by Eli Goldratt, critical chain is a methodology for planning, executing and managing projects. Goldratt first introduced it in 1997 through his book Critical Chain, one of the seminal tomes on the theory of constraints. He developed this method in response to his understanding that poor performance resulted in cost overruns, frequently missed deadlines and underdelivery of scope.
Goldratt theorized that traditional tools used to manage projects did not effectively address the counterproductive human behaviors that are incentivized by embedding safety time into task durations as a buffer between tasks. These time buffers tend to encourage project participants, also known as resources, to push project actions off until the last moment before a deadline, sometimes making it impossible to recover from a delay or problem. Similarly, these buffers—also known as project safety buffers—discourage project participants from reporting early delivery of tasks that could significantly shorten future deadlines. The result of these behaviors is that while delays are always passed on to affect the whole project, benefits from early task completions are rarely realized. Rather than expect the same results for project resources, regardless of project size and scope, Goldratt saw an opportunity to extend existing project management frameworks to capitalize on early task completion.
Why Use Critical Chain?
Between 2009 and 2013, national-scale residential solar businesses, such as SolarCity, Sunrun, Sungevity and Vivint Solar, attempted to standardize project delivery to support their growth ambitions. They developed and used project management software while predetermining acceptable site and project criteria. Unfortunately, solar projects—regardless of market sector—have more in common with custom home building or remodeling than with factory work in that each project is custom.
Timelines for some solar project tasks—such as receiving permit or interconnection approval or addressing special site conditions—can add anywhere from a week to a year to a project schedule. Delays can easily compound or threaten the budget. So how do project developers balance project payment milestones and task delivery schedules to minimize risks?
This is where a project management technique such as critical chain comes in. To illustrate, let us step through the mechanics of critical chain project management.
Mechanics of Critical Chain
Unlike traditional project management methodologies that evaluate project progress in terms of individual tasks, critical chain manages projects to your resource limitations, such as availability of engineers to produce permit-related documents, and substitutes individual task completion buffers with one large buffer at the end of the project. When a project experiences a delay, project managers can focus on the specific activities that most impact the buffer and intervene accordingly.
In this framework, the project management resources are people and their capacity limits for completing tasks. The role of the project manager is not to name tasks and line them up, but rather to understand resource capabilities and project completion requirements. Critical chain project management incorporates three types of buffers: an overall project buffer, feeding buffers and resource buffers.
Project buffer. Project managers use the overall project buffer at the end of the project between the final task and the completion date to monitor project timeliness, as shown in Figure 3. Ideally, delays on the longest chain of dependent tasks will eat away at the buffer but leave the completion date unchanged. Unlike a traditionally managed project that has time buffers built in to each task, the total project buffer allocated in a critical chain will be smaller than the sum of all individual task buffer times of that traditionally managed project.
Feeding buffer. Feeding paths are those tasks that are not on the critical chain. Delays from feeding paths can impact the project by delaying a subsequent task on the critical chain. To protect progress, project managers insert feeding buffers between the last task on a feeding path and the critical chain. The total duration of the feeding buffer should be half the size of the sum of safety time taken out of the feeding path.
Resource buffer. Project managers use resource buffers to account for the human elements of a project. This buffer addresses who is available to perform specific tasks, as well as any constraints they have in terms of skills, capabilities and tools. Resource constraints are the hardest project variable for project managers to change and are often what prevents completion of a project in the shortest possible time. Therefore, project managers must generally elevate resource management above other concerns and plan to take these constraints into account.
Once project managers set these buffers, they can monitor project health and timeliness by watching the buffer consumption rate rather than individual task completion to keep everything moving forward on schedule. This keeps a project manager from unnecessarily riding herd on feeding tasks that are running late but that will not significantly impact the critical chain.
Implementing Critical Chain
The following steps in critical chain project management illustrate how to apply these procedures to solar projects.
Step 1: Plan and estimate. Start with a project plan or work breakdown structure that identifies all the tasks necessary for project completion. Work through the tasks backward, starting from a completion date, with each task starting as late as possible. Assign two durations to each task: The first date is your best guess at the duration, with a 50% probability of completion; the second date is your conservative estimate of the duration, with a 90% probability of completion.
Step 2: Allocate resources. Assign resources to each task by deciding who will do what. To balance the demand for resources with the supply, employ resource leveling across the plan by analyzing previous projects to understand the production capabilities of installation technicians, designers, engineers, administrative staff and other project actors. Develop a project schedule using the most aggressively short durations possible. The critical chain improves on the critical path method because the critical chain is the longest sequence of tasks, with resource constraints applied, leading from the beginning to the end of the project. You can break out other tasks into the feeding paths leading into that critical chain. The difference between the 90% and 50% durations you determined earlier becomes the buffer at the end of the project.
Step 3: Start the relay race. Critical chain project execution is like a relay race. As the runner with the baton approaches the person responsible for the next leg of the race, that next person takes off and both runners run side by side until they reach the handoff point. Similarly, the goal of critical chain project management is to eliminate waiting times between tasks by allowing the next person in the critical chain to see when the previous task is coming to completion and his or her task is about to start. Balancing the workload of resources is where the art and science of resource leveling comes in. Project execution is also a potential source of work culture conflicts. In project environments with many interruptions, resources are often under pressure to multitask. Critical chain discourages multitasking or any delay to the start of a task.
Step 4: Monitor progress. In a critical chain, individual tasks will always vary in duration from the 50% estimate. Pressuring resources to complete every task on time is an exercise in futility. There is no need for the perfect to be the enemy of the good. Instead, project managers and project actors should monitor the buffers agreed upon during the planning stage. As long as the remaining buffers are adequate, everyone is doing a good enough job of keeping the project on schedule. Often, this “good enough” aspect of resource-based critical chain project management is the most difficult mental hurdle for project managers more familiar with task-based Gantt chart project management.
Organizations generally rely on software to efficiently plan, execute and track progress along the critical chain. Rather than tracking time elapsed from the start of a task to completion, critical chain software solutions include work queues and dashboards that monitor the rate of buffer consumption. If the rate of buffer consumption is low, the project is on target. If the rate of consumption is trending higher, then project stakeholders need to take corrective actions or develop a recovery plan to recoup the lost buffer. These buffer-rate consumption data allow project managers to set the appropriate project pace and conduct requisite resource-leveling activities.
Critical chain is a powerful way to reinvent your approach to project management to deliver projects on scope, ahead of schedule and under budget. Note, however, that you will have to make some important organizational changes to switch your project management methodology. To switch to critical chain project management, you must start by tracking project constraints and analyzing the effectiveness of your resources across a variety of tasks. You also need to evaluate the software you are using and take a fresh look at your planning process and project management style. Most important, you have to ensure that your company culture prioritizes resource dependencies along the critical chain, ignoring all distractions. Without these elements, it is easy to revert back to task-based project management styles.
—Pamela Cargill / Chaolysti / Alameda, CA / chaolysti.com