Being the Best Does Not Always Result in the Best Outcomes

During the last 15 years of my 38-year career, I led a team of electrical and instrument professionals who were involved in the design, installation, commissioning and startup of a number of major Upstream oil & gas projects. For the experienced professionals in the group, they were selected because they were considered the best in their field within our Upstream business unit. Many of these experienced professionals participated in industry committees, gave speeches at industry events and worked with many of our suppliers in the development of new industry technologies. For the new hires in the group, we always hired the very best. 

All of these projects fell into the category of mega-projects. They each cost multi-billion dollars, involved multiple engineering contractors in different locations, had multiple fabrication sites around the globe, utilized packaged equipment from suppliers around the globe and took 4+ years to complete. 

In the early years, we experienced all of the historical problems every company in the oil & gas industry and quite frankly, all process industries had experienced with delivering power and automation systems for major projects. This included numerous changes throughout the project execution, numerous change orders, increased costs, late completion of engineering, increased engineering contractor resources to implement the changes, late delivery of procured equipment, late delivery of vendor data, late modifications to equipment already delivered to the field, late completion of construction of the I&E systems and extended commissioning schedules with increased need for commissioning personnel. Overall project schedules were still being met but extensive extraordinary measures were necessary on every project to achieve these schedules. The extent of each of these problems varied between projects but every problem occurred on every project. The I&E deliverables and activities always defined the project critical path. These outcomes were not unique to the oil & gas industry. These outcomes were common to all process industries. 

Typically, blame for these problems were placed on the internal I&E staff, the engineering contractor's I&E staff, the automation systems suppliers and/or the power equipment suppliers. Historical industry procedures were being used for every aspect of these projects, so it was assumed the problems had to be caused by people either in their skills or in their execution of the procedures. 

From an I&E perspective, we could typically blame the process engineers, the loss prevention engineers and the mechanical engineers for the frequent additions of process instruments, safety instruments, process motors throughout the project execution. From the perspective of these engineers, they were simply doing what they have always done. It was very common on all projects by all companies that instruments could be added right up until startup. 

What we ultimately realized was that the problems were generally not caused by people. We could have had the very best internal resources working on a project with the very best engineering contractors and the very best suppliers and the problems would still occur. We spent a lot of time identifying the best internal resources for every project. We spent a lot of time selecting the best engineering contractors and the best suppliers. We even focused on the historical processes to perfect them and execute them properly. With all this, the problems still occurred. 

The problems were not caused by the people involved but by the processes being used. Processes that were developed for smaller, less complex projects using a single local engineering contractor with local suppliers of packaged equipment simply will not work for complex mega-projects using multiple remote engineering contractors with multiple remote fabrication sites and with equipment purchased from around the globe. It was the processes that had to change and not the people. I would even argue that if you have the very best resources using the outdated historical processes, the problems created by the historical processes can be magnified simply because the historical processes are being followed in detail. 

Just as an example, the way automation systems were historical designed included marshalling cabinets to properly connect specific field instrument types to specific control system input devices. This also resulted in a very large number of field junction boxes or building mounted cabinets that were each uniquely designed to accommodate the thousands of instruments on a typical project. No junction box or cabinet was the same in the design details. For a typical mega-project, there could be 50,000 I/O with 400-500 field junction boxes and internal cabinets with each junction box/cabinet being uniquely designed so that every time a single instrument is added, literally hundreds of drawings are impacted. On one project, we went through the complete alphabet for revisions to the cable schedule and almost went through the complete alphabet in double letters. This was almost 50 re-issues of the cable schedule. With the hundreds of instruments that are added on a typical project after the initial design right up until startup, the number of changes to drawings were significant and, on many projects, it was almost impossible for the I&E teams to keep up with the changes. The way automation systems were designed had to change.

We described the problems we were experiencing on every project with all of our key suppliers, and they all developed the solutions that resolved every single problem. They developed universal I/O that eliminated the need for marshalling cabinets. They standardized on the field junction boxes so that no field junction box had to be uniquely designed and could be mass produced and ordered very early in the project even before the design was complete. They moved the I/O connection devices to the field junction boxes eliminating the need for I/O cabinets. They developed virtual FATs that could be performed without the target hardware and could be performed anywhere in the cloud. This also eliminated the need to stage the complete system at a staging location. The system components would be mass produced and sent directly to the field and when the program was downloaded, everything just worked. They developed automatic commissioning which could automatically occur when the instrument is connected to the control system. Finally, vendor standard solutions were developed with each of our automation and power system suppliers which significantly reduced the effort and time to procure package equipment. It also significantly reduced the cost of the equipment and improved the quality of the equipment.

It is not enough to just have the very best resources, contractors and suppliers. You must also have the very best processes and procedures for the specific project you are executing. Problems are not always caused by people but by the processes used. Please do not assume that when project problems developed, it is caused the the project participants.

I do not think these improvements are unique to I&E. If companies are still developing very detailed specifications with ANY non-vendor standard features, they are increasing the cost of their procured equipment by up to 30% and forcing procedures that extend equipment deliveries and consume unnecessary resources. This standard equipment should also have standard vendor drawings that could be downloaded when the equipment is purchased. There should be published PI&Ds that are utilized by all process industries. There should be standard published designs for separations systems, dehydrating systems, etc. If a company is adopting standard solutions, it should apply this approach to every aspect of the project and not to just limited components and solutions. As I have mentioned in other articles, project delivery should be about execution and not design development either in equipment or systems. 

I would suggest that every historical project delivery processes should be challenged and dramatically changed considering new tools available. Simply "improving" historical processes is no longer enough.