May 2022

Plant Turnarounds and Project Management

Managing complexity in engineering and construction projects

It is widely accepted that new management methods are needed to curb the frequent delays and cost overruns observed in engineering and construction projects.

Corrales, J., Veiga, J. P., Tecnicas Reunidas

It is widely accepted that new management methods are needed to curb the frequent delays and cost overruns observed in engineering and construction projects. New approaches, such as the Last Planner system of production control and advanced construction packages, can help to increase construction efficiencies; however, none of them address the root causes of problems, which is the difficulty of managing the inherent complexities of large engineering and construction projects.

This article highlights the deficiencies of today’s working practices and proposes a practical method to manage complexity.

The failure of conventional project management

Numerous surveys conducted in recent years have shown that conventional management methodologies are incapable of avoiding schedule and cost overruns in engineering and construction projects. A macro survey conducted by Ernst & Young (EY) on 365 oil and gas projects revealed that 65% had long delays, 78% had suffered significant cost deviations, and the average budget overrun was 53%.1

For some time, it was thought that tools such as 3D computer-aided design and building information modeling would solve the problem; however, while they have contributed to substantial improvements in safety and technical quality, their impact on cost and schedule has not been very significant.

The first attempt to abandon the traditional project management approach was made by Glenn Ballard and Gregory Howell in the 1990s when they developed the Last Planner methodology. In essence, they applied lean management concepts developed by Toyota. Inspired by the “pull” principle—which states that things must be done “when and as required by the last links in the project chain”—and under the conviction that the detailed initial schedule is always totally wrong, they proposed to start with a simple master schedule and to leave the detailed planning to the construction foremen.

While the Last Planner methodology is widely used by building and civil works contractors, it has had little success on oil and gas projects. The reason could be that oil and gas projects involve multiple contractors—therefore, it is very difficult to obtain timely feedback from construction foremen.

Information technology (IT) and new product development projects have already abandoned the traditional management approach. A group of software developers launched the Agile Manifesto for software development in 2001, which implies a total departure from conventional project practices. Today, these software developers use methodologies such as Scrum, which does not have an overall schedule. Instead, the project is divided into sequential steps of no more than 4 wk. The final scope is frequently redefined, and the contractual relations are totally flexible. This approach has been so successful that many wonder why our industry does not apply it to engineering, procurement and construction (EPC) projects. The primary reason is that oil and gas EPC projects are drastically different. The problem encountered in IT projects is that the objective is not entirely defined, and the design specifications are few or totally nonexistent. Conversely, in EPC projects, the objective for the final product is clear from the beginning, and the volume of design codes and specifications is overwhelming.

The last divergence from conventional project management is the Advanced Work Packaging (AWP) methodology, which was co-developed by the Construction Industry Institute (CII) and the Construction Owners Association of Alberta (COAA) in Canada. These organizations found that the average hands-on-tools time for construction personnel is below 37%, and that the rest of their time is wasted in document/material searches or from idleness. To try to solve this challenge, the researchers proposed a procedure based on dividing the construction work into packages and transferring the responsibility for planning to the field. Instead of leaving it to the foremen, as the Last Planner system does, they employ specialized construction planners. Each week, the planners prepare the packages that are fully ready for execution and deliver them to the foremen. Additionally, to enforce a construction-driven design and an orderly delivery of information and materials, engineering disciplines are required to prepare a separate information package for each of the construction packages.

According to the CII, this method can lead to improvements of up to 25% in time-on-tools efforts for construction labor. Depending on the type of project and geographic area, this would represent savings of approximately 5% of the total project cost. This is an impressive savings, but it is not enough to drive away the two-figure deviations found in many EPC projects.

The authors’ thesis is that the main cause of cost and schedule overruns is the intrinsic complexity of EPC projects; therefore, the problem will only be solved when the industry finds a way to effectively manage that complexity.

The roots of the problem

The causes of EPC deviations most frequently mentioned in literature include the following:

  • Client design changes
  • Low construction productivity
  • Aggressive bidding
  • Poor project planning and scheduling
  • Financing problems
  • Slow decision-making
  • Acts of God.

If these deviations are analyzed, the underlying cause is the complexity of EPC projects. In theory, client design changes should not be harmful because the contractor will be compensated, and the client only makes them if the benefit/cost ratio is positive. If changes end up being a problem, it is because their impact turns out to be worse than expected. The reason this happens so often is that project complexity makes it difficult to foresee the real impact of the changes. It also makes it very difficult to reach an equitable agreement between the parties on the costs of the change.

The second deviation—low construction productivity—is not due to lack of experience of construction personnel, but to problems caused by delays in the delivery of drawings and installation materials. The complexity of the project should be the only reason for highly experienced engineering and procurement teams to end up doing a bad job.

The third cause is aggressive bidding. Market pressures can often force contractors to provide low prices, but the complexity of EPC projects means that, even unintentionally, bids can easily be unrealistically high or dangerously low.

Complexity plays an important role in all the deviations, including Acts of God, because risk analysis and mitigation measures are always carried out before starting a project. If they prove to be insufficient, it is because the complexity of the project makes it very difficult to gauge the consequences of the acts.

While there are other factors that can also lead to project failure, complexity is by far the most common and most difficult to manage. In addition, complexity is aggravated by the urgency of capital projects. Project lead times are excessively long for an increasingly dynamic economy, and this makes urgency present at every step of the project. This urgency exacerbates the consequences of complexity because, in addition to limiting the time available to try to untangle the complexities of a project, urgency often forces the design to move forward based on assumptions rather than on firm data.

Academic research supports the conclusions drawn in this article. After a systematic analysis of 86 research papers on the causes of project failures, Denicol et al.2 found that the dimensions that make megaprojects so difficult to manage include size, uncertainty, complexity, urgency and institutional structure.

There are numerous research papers on the origin of project complexity, but, in the case of EPC projects, one only needs to see the detailed critical path method (CPM) and program evaluation and review technique (PERT) network to understand where this complexity comes from. A medium-sized EPC project can easily have more than 20,000 tasks, with a much higher number of interactions or dependencies among them. Like trees that obscure the forest, complexities blur and hide problems, and, without a clear vision, it is very difficult to manage them.

Prieto3 said that, in addition to obscuring the problems, complexity amplifies even the smallest and seemingly improbable risks. In his book, The Improbability Principle—Why Coincidences, Miracles and Rare Events Happen Every Day, David J. Hand mathematically describes the amplification mechanisms, such as Lorenz’s butterfly effect. The following will show that the size and complexity of this maze of multidisciplinary tasks and interactions are further aggravated by the limitations of the tools used to visualize this maze.

The schedule problem

Scheduling is a key tool for efficiently planning and controlling the project’s execution. One might think that the schedule’s function is limited to controlling execution times, but scheduling also has a strong impact on many other factors—such as spent worker hours, the amount of rework required, construction efficiency, the necessary volume of surplus materials, and many other cost elements that are affected by the schedule design. The following are several examples:

  • If the design of cable routings is started with immature information, designers will have difficulties obtaining the data they need and spent worker hours will increase. Furthermore, they may have to work on assumptions, which may, in turn, lead to a wave of redesigns in downstream disciplines—including construction reworks—when the correct data arrives.
  • If construction subcontractors are mobilized too early, they will suffer from drawing shortages, leading to a drop in efficiency.
  • If time frames assigned to procurement and subcontracting tasks are too short, chances are that there will not be enough time to get the most competitive prices.
  • If the initial schedule is too optimistic, the project will accumulate delays, often leading to staff demotivation.

The schedule is, therefore, the most powerful tool available to project management. The budget is an essential reference for cost control, but it is only a reference and not a mechanism that can act directly on costs.

There are very powerful schedule management tools on the market. While they are designed to display a myriad of activities, with the ability to drill down into the details, they do not provide an all-encompassing vision of project structure. These tools cannot produce clear summaries capable of communicating the complex task networks and the roles of the interactions between them.

Managing a project without a clear overall vision of its structure is like driving a car in the fog. There must be a clear overall picture to enable the EPC contractor to detect risks and opportunities, measure the consequences of decisions on downstream activities, and choose the best implementation strategies.

Mapping of project interactions

There are few solutions available to obtain a clear overview needed to manage large, complex projects. If these solutions are simple tools and are not complicated systems that add more complexity to the already overloaded project management environment, then advanced analytics can help.

One possibility is the use of simplified schedule representations. By grouping common tasks, one can draw a map or a network showing the essential groups and their main interactions. A simplified project interactions map is shown in FIG. 1. This mapping should be sufficiently detailed to shed light on execution strategy decisions—including obstacles to overcome to advance equipment inquiries, prioritization of equipment procurement, identification of areas where value engineering exercises should be performed, modularization, and conditions for mobilization of construction subcontractors, among others.

FIG. 1. Simplified project interactions map.
FIG. 1. Simplified project interactions map.

The first time the map is prepared, this task can be very laborious, requiring the involvement of all project disciplines, especially construction. However, once the first one has been drawn up, it is very easy to adapt it to each new project.

During the launching of a new job, the map is used by the management team to decide the primary execution strategies. Once these strategies have been determined and reflected in the map, it can be used as a guide for schedulers in the development of the initial detailed schedule. In addition to improving the reliability of the schedule, this map drastically reduces the time required to complete it.

After the detailed schedule has been issued, daily work is controlled by it. However, it is advisable to keep the map alive and updated throughout the life of the project, both for its usefulness for decision-making and also to incorporate lessons learned that may be useful in future occasions.

The authors have called these documents “interaction” maps to stress the importance of interactions or dependencies between tasks. A mere change in a dependency can transform a project. For example, if starting the civil design of a process area after receiving 90% of the vendor’s drawings (projects are typically started with 60% of the vendor’s drawings), the project will be completely different. Civil works may start earlier, but there will be more drawing revisions, more site reworks and more change orders.

While tasks and durations are clearly shown in multiple project documents, dependencies are practically ignored. Project schedulers use them to build the PERT-CPM network, but they do not bother to give them much publicity because they realize that few people will take the time to review them. Moreover, the arrows used to represent them tend to get very intricate, and there is no easy way to reflect them in project reports. Since dependencies receive little attention, there is very little knowledge about their impact on project results. This is a major problem because the art of planning is the art of wisely selecting dependencies. HP


  1.  EY, “Spotlight on oil and gas megaprojects,” 2014, online:
  2. Denicol, J., A. Davies and I. Krystallis, “What are the causes and cures of poor project megaproject performance? A systematic literature review and research agenda,” Project Management Journal, February 2020.
  3. Prieto, R., “Theory of management of large complex projects,” Construction Management Association of America, November 2015.

The Authors

Related Articles

From the Archive



{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}