New-Plant Project Management

Plant build-out—greenfield or brownfield—requires advanced project management and controls. But it doesn’t have to pit EPCs against MACs. Here are some of the considerations.

Construction of new process facilities: Projects range from line upgrades that cost less than $1 million to huge new campuses costing more than $4 billion. The geographic range is equally wide, with larger greenfield projects more likely in developing regions, and brownfield plant renovation projects more likely in North America.

But despite the wide range, they have one thing in common. They are systematic, collaborative efforts drawing on the expertise and experience of multiple disciplines, from accountants pushing essentially weightless electrons to construction navvies wielding tons of concrete.

In the end, both brownfield and greenfield plant projects are embodiments of systems for manufacture. The resulting plant space is (most people hope) a well-organized machine for production. So, too, is the project that results in the new space. At its most abstract, plant projects are also well-organized machines—systems that include the project owner, project managers for engineering, procurement and construction (EPCs) and/or main automation contractors (MACs).

As with any large project involving numerous professionals, an EPC/MAC project can present challenges to plant owners, engineers and operators. But everyone can work together to ensure a new or revamped plant meets both operational and business objectives.

An EPC/MAC project has much in common with large-scale technical projects in general. “FEE [front-end engineering] and preliminary planning are critical,” says Cameron Thrutchley, director of strategy at Honeywell Process Solutions, a Phoenix-based MAC. He cites front-end loading (FEL) levels as developed by Independent Project Analysis, Inc., an institute that provides education related to capital projects:

• FEL1, evaluation of business needs, where you “set your vision,” Thrutchley explains;
• FEL2, development of scope—“your philosophy around automation, safety, networks, all the issues”; and
• FEL3, authorizing the right project via “a detailed functional spec, one that’s integrated with the vision determined in the first stage.”

“The key moments for any EPC or MAC project comes at the beginning,” agrees Mark Freedman, senior partner with The Boston Consulting Group (BCG), and global leader of its Industrial Goods practice. “It’s essential to define an execution plan and stick to it.”

Still, while the basics devolve into tried-and-true capital project parameters, the details—and there are many of them—combine into a very special kind of project: one that requires an unusual amount of back-and-forth education among the players (because no single discipline ultimately knows much about the work the others do), and an unusual amount of collaboration (because nearly every detail is in one way or another tied to all the other details).

Because everything is under the gun—the plant’s safety, economic payback and the need for quality production at target rates—Freedman emphasizes one extremely important early step: “Put five to ten main risks down on paper—including how to mitigate them. It’s surprising how often this is not done.”

91.7 percent late

What happens without front-end engineering and design? A 2004 Honeywell research project involving more than 1,000 customers and project owners revealed that 91.7 percent of facility building or rebuilding projects finished late. Thrutchley adds, “And once we looked at the details, the situation was worse than this statistic reveals. There were the usual issues of budget and schedule, but there were also troubles during handoff to operations. Worse, many of the as-commissioned lines didn’t hit targeted production rates, sometimes for years after startup.”

That last issue is a primary area of concern. Thrutchley cites the conclusion of a study by one of Honeywell’s major oil and gas customers: a given facility will never reach its projected return on investment (ROI) objectives if it has failed to reach its production targets within a year.

ON THE WEB: 4 Questions—Project Management
Learn about small-scale project management in May 2011’s 4 Questions with John Person, vp, Tangent Engineering. Visit

Thrutchley blames what he calls the traditional model for many of the problems—a model that sees an owner contracting with an EPC, who then throws contracts over the wall to an MAC for the automation portion.

“EPCs are world-class at plant design and build-out,” he says, “but the increasing proliferation of automation systems and controls, and the increasingly finer granularity of sensor and operational data, all add up to a situation where a typical EPC just can’t keep up. If the EPC works from a traditional, vendor-transactional point of view, an automation source becomes just a ‘vendor’ selected for cost alone. ‘Give me a DCS… or SIS… or field device in a box, three bids and a buy.’ But it just doesn’t work, 97 percent of the time. Important automation opportunities and crucial integration points remain unseen, delays and engineering changes proliferate, and the result is usually a suboptimal system.”

An integrated approach

Thrutchley urges an integrated methodology, where the EPC and the MAC act as peers. “A MAC has to be a strategic partner and consultant,” he says. “Nobody else knows the depth and breadth of the products available—or as much about installation and integration needs. When the EPC and the MAC are peers, you see the best of both worlds. The EPC handles plant design, procurement and build-out, while the MAC oversees an automation design and delivery effort that ideally is focused on both operations and business objectives.”

“We see two essentially different kinds of projects,” says Chuck Egasti, general manager of systems operations for Yokogawa Corp. of America, a MAC based in Sugar Land, Texas. “One is the brownfield type, where the MAC acts as a turnkey supplier with no EPC involvement. Most U.S. projects fall in this category. The second is greenfield projects, where there can be multiple EPCs. We think it’s important that there be a single MAC, so that the automation systems benefit from a cohesive look and feel.”

Egasti says project owners among top-tier companies usually have a good handle on what they want. “They’re big enough to have resources on board to facilitate project development. Below that level, those resources are not available internally. You’re likely to find a group that is actively exploring solutions with outside input. They tell us, in effect, ‘Help us get to where we want.’ ”

It becomes an education process, an iterative process, Egasti says. “‘Here’s one thing you can do,’ we say. Then they say, ‘Great—now that we see that, what about thus-and-such?’ We come back with a further solution, they approve or they hesitate to invest. We explain more—and so on. A lot of education is just part of the process.”
Bernie Rieskamp, control systems and instrumentation discipline specialist for Jedson Engineering, an EPC firm in Milford, Ohio says, “Projects tend to be developmental and iterative in nature. It’s important to gather all the stakeholders and ask all the questions before scope development. This can involve more time than you’d think. It’s not uncommon to find functional groups within a company that are not 100 percent aligned with each other.”

Jedson, which specializes in half million to $150 million EPC projects in chemicals, pharmaceuticals, health and beauty, and similar processes, pushes for upfront definition of needs. “You need to get all the details up front, and by that I mean all the control narratives, cycle times, process safety implications, programming, simulation—everything,” Rieskamp says. “If you don’t spend time in the details, then you’re going to spend more time and a lot more money on change orders, and in startup and commission.”

“Take your choice,” he adds. “Face a huge punch list, or reap the rewards of a line start that’s making sellable product within hours.”

Underlying tension

Simmering below the surface of many projects is a fundamental point of discussion, if not out-and-out conflict. As we have seen, MACs tend to want to either lead or act as project peers with EPCs. On the other hand, EPCs tend to want to lead. Jedson, for example, prefers projects which are completely under its control, though with plenty of input opportunity if an MAC is involved, says Rieskamp.

“The EPC always wants to be in charge,” says Eric Fidler, director of oil and gas business for Rockwell Automation, a Milwaukee-based MAC. “Unfortunately, the typical EPC execution model is to optimize cost in silos. That doesn’t necessarily drive to an overall system cost—the cheapest motor control center may be suboptimal compared to an intelligent one, for example, where you can save money both on interwiring and commissioning, and later during ongoing operations. If you silo things, you run the risk of everyone trying to optimize only in their own bucket. You miss systems savings if the EPC and MAC are not around the table as peers.”

“It’s surprising how often the interfaces between the project owners and all the contractors are left undefined,” says BCG’s Freedman. “It’s better to scope the project precisely, and make sure that incentive clauses are aligned across all the players. And, while you hope that friction never occurs, if claims and counter-claims do happen, you have to avoid making them an endless source of discord. A common scenario sees one group that is driven to finish quickly, maybe without the right quality standards, and another that obsesses about quality to the point of delay. You need to negotiate an acceptable level of quality before the project rolls.”

Owners and their customers are the key, says Rockwell’s Fidler. “This includes not only customers in the usual sense—buyers of product—but the owner’s internal customers as well, he says. “Projects often begin with an enterprise’s capital project group. They sometimes don’t realize that operations personnel are their customers—no plant will work well if operations has not taken ownership, or perceives the final outcome as unmanageable. The same applies if facilities people, corporate standards, or instrumentation personnel are left out of the planning.”

Bill Robertson, manager of North American sales services for Emerson Process Management, a MAC based in Austin, Texas, takes this further, into what he calls the nuances: “The project owner ultimately wants what’s best for the plant in the long run, and that depends on production personnel, right down to operators. After all, operators have to see specific data and trends to control processes, and they hew to closely defined ways to execute sequences. All this affects decisions around automation systems and equipment. Unfortunately, in the worst case, the EPC wants only what is cheapest. Only with direct contact with an MAC can the owner directly influence decisions around the best means to the defined operational objectives.”

Managing all the horses

William Mao, vice president of the U.S. oil, gas and petrochemical business unit for ABB, a MAC with global headquarters in Zurich, emphasizes that the overall project universe is large and varied. “Most of our automation contracts are spearheaded by the end user,” he says, “because many are global contracts by major producers. If the project is greater than four to five billion dollars, EPCs usually are in the picture. We’ve seen as many as 22 EPCs involved, with skill sets and knowledge levels all over the place, from woeful to world-class.”

ABB’s solution is to build multiple touchpoints into projects, taking advantage of communication channels to update and inform. “We educate when it’s appropriate,” Mao says. “We also work directly with highly specialized EPCs, who educate us. We push for collaboration at all levels, and we work from internally-published best practices across all the layers of a project.”

The final challenges to any plant project is handing off to production. “Ultimately, your objective is to build a facility that from day one is business-ready,” says Honeywell’s Thrutchley. A chain of positive project attributes, he says, sets the stage: “Strong leadership, especially top-level commitment from the owner, an automation supplier with proven methodologies, adequate front-end loading, clear project scope, detailed specifications—these all contribute.”

>> September 2011, Industrial Megaprojects: A Resource
To read the feature article, visit

Boston Consulting Group (
Emerson Process Management (
Honeywell Process Solutions (
Independent Project Analysis, Inc. (
Jedsen Engineering (
Rockwell Automation (
Yokogawa (

 Dave Gehman is a Contributing Editor with Automation World.

More in Home