This article is the second in a series in which Jim Carter examines the lessons project owners can learn from the ways nuclear construction projects have gone wrong in the past. He shares observations from research and years of experience in nuclear power to help deliver successful deployments of all kinds. You can read the first post here.
In my previous blog post, I focused on scoping and engineering – how the nuclear industry can incorporate improved tools and techniques and current best practices to start a project off on the right foot. This time, I take a look at ways to better keep projects on track once they’ve begun.
First, I’ll go through some of the challenges we’ve faced in past deployments. Then I’ll talk about how we can improve on those situations for future projects.
Previous Challenges
Estimating and Scheduling. Historically, inaccurate schedules and budgets begat inaccurate and ineffective execution, which in turn contributed to project cost overruns and schedule delays.
Incomplete designs created high uncertainty. Estimates and schedules were not uniform, as various input sources had different degrees of expertise, rigor and optimism. Reviews and vetting were often perfunctory, sometimes performed by “interested parties” instead of independent experts.
The resultant plans often had limited value in managing a project and were difficult to accept by those responsible for performing the work. Schedules were frequently organization- or discipline-based, within silos that challenged effective interfacing.
The scheduling and estimating tools were incapable of frequent modification to sustain a valid plan. As engineering and design matured, costs escalated and schedules slipped.
Contractor change orders became extensive and often were aggregated into one mega budget increase request and hastily approved. Stakeholder confidence waned, and “righting the ship” became increasingly difficult.
Risk Management and Contingency. During the 70s and 80s, programmatic risk identification and proactive mitigation efforts were virtually non-existent.
Design risks affecting plant operation and safety were addressed in great detail, as one would expect for a nuclear plant. Project managers, however, dealt with commercial risks as they arose and suffered unanticipated schedule and cost impacts.
There were no preemptive mitigation efforts, and a reactive corrective response was triggered when the risk was realized. Solutions developed on the spot often carried painful schedule and cost impacts.
Contingency for estimates and schedules was usually informally embedded in estimate line items and schedule durations. This was not visible for objective review and vetting; was ad hoc; and varied with the individual sources.
Reporting. Early nuclear projects had limited tools for effective updating and progress reporting. Project schedules, rudimentary by today’s standards, were qualitatively “statused” by estimating the percent complete of the activities.
Given the time and labor to obtain and then process such input, schedule status reports were outdated by the time they were issued. They were subject to considerable uncertainty. While these schedule reports were of general use, they lacked credibility. They were often discarded in favor of subjective reports based on observation and self-generated lists. Progress was often measured by milestone achievement.
There was usually a high reluctance to convey bad news. Contractors worked to correct problems without reporting them, in hopes that additional effort would cause resolution without fanfare. This occasionally worked, but when it did not, optimum solutions were usually unavailable due to the passing of time.
Technical quality was subject to independent inspection and verification. Therefore, it rarely suffered the information suppression associated with schedule problems. No such program existed for commercial quality.
Commissioning. In the past, test program activities were rarely scheduled in initial plans by knowledgeable commissioning professionals, who usually arrived on a project very late in the game. Standardization was totally absent. Commissioning schedules created by unqualified professionals were essentially placeholders with surmised bulk allowances for testing and turnover.
There were no detailed interface ties between construction and test program activities. Without that integration, construction sequencing could not accommodate test program requirements without significant changes. This resulted in significant changes to schedule and budget as well.
Procurement. In the past, supply chain activities were impacted on the front end by the timely availability of engineering and design deliverables needed for specifying the material and components.
On the back end, supply chain must know when site delivery is required to support construction and avoid excess site warehousing and carrying costs. This clearly challenges the procurement process.
Suggestions for New Nuclear Deployment
Using tools and best practices that are available today, the industry can reduce the impact of these previous challenges and improve project performance.
Fully Integrated Project Schedules. The value of a fully integrated detailed project schedule cannot be overstated. It is the backbone of project management.
Schedule development and integration should extend from the early concept stage through detailed engineering, procurement, construction, commissioning and integrated plant startup. Early integrated schedules may not be perfect, but they are valuable for establishing priorities for engineering and procurement activities.
Engineering teams must know when their deliverables are required to support later activities such as other engineering disciplines, procurement solicitations and orders, construction work packages, test program procedures, O&M training and many other activities.
The Association for the Advancement of Cost Engineering (AACE) has developed useful guidelines for identifying “expected” estimate and schedule accuracy ranges based on the maturity level of project definition deliverables.
Risk Management and Contingency. In general, state-of-the-art risk management processes involve a comprehensive effort to identify risks to project success. These are defined by describing the event, the cause, and the impact.
The probability that the risk will occur is established by experience. Values for cost and schedule impacts are estimated within a set of prescribed ranges. Risks can then be scored, and owners can focus mitigation strategies on those with high scores.
To create contingency, more quantification is necessary. Three point values reflecting low, mid and high likelihood can be applied to each risk cost and schedule impact. A Monte Carlo analysis is then performed to establish contingency with the desired level of confidence.
Similar ranging and Monte Carlo analysis is done for elements of the estimate. Combined, an aggregate value of project contingency can be established.
Reporting. Modern tools and techniques make effective schedule and budget development and tracking easier. However, they are only as good as the input and management’s response. Estimate and schedule development must go hand in hand. Nuclear stakeholders should devote significant attention to developing the critical initial project schedules and budgets.
For today’s plants, the benefits of standardization can be significant when it is effectively applied to schedules and budgets. Nuclear developers would be well served to monitor actual schedule and budget performance on each project and update the “standard” schedule and budget packages based on field actuals when appropriate.
Commissioning. The interface between construction and testing is a difficult one. After years of operating in field construction conditions, the project must shift to test program conditions with live switchgear and pressurized/heated pipes; and ultimately to operating nuclear plant conditions. Effectively preparing for this challenge must occur during the early project planning stages.
Nuclear developers would be well served to take advantage of the standard design concept and establish a standard test program that deals with all aspects of commissioning from construction tests (such as pressure testing) to preoperational and integrated systems testing.
Technical test procedures and programmatic matters such as provisions for turnover boundaries and turnover package requirements should also be addressed.
Procurement. Again, the value of a fully integrated, total project schedule cannot be overstated. As addressed above, engineering teams must know when their deliverables are required to support later activities.
Schedule support is not the only consideration. Engineering documents must be consistent with procurement requirements and supplier capabilities.
Nuclear stakeholders should ensure that processes and procedures are current and reflect the needs of all parties, including the commissioning and operations teams.
Jim writes more extensively about this topic in his paper, “Small modular reactor deployment: Learning from the past and the present”, which was published in The Electricity Journal. Download the paper here.
