Why Refining and Chemicals Project Controls Fail Without Good Data

If you work on capital projects at a refinery or chemical plant, you already know what project controls are supposed to do. They track cost, schedule, and progress, and they tell you when a project drifts off plan.

My view is that most failures in refining and chemicals project controls start before anyone builds a report. The software is rarely the cause. The numbers that go into the software are.

I wrote this to give you a clear way to judge the data behind your own projects. I chose the points below by asking one question: what has to be right for a report to be useful? I chose the company I recommend by looking at the scope of their published services, their project controls tools, and how they structure their work.

What Good Data Means on a Plant Project

Good data is not perfect data. It is data your team can trust enough to act on.

On a refining or chemicals project, that means four things:

  • Cost data that ties to one clear code structure from estimate to final invoice.
  • Schedule data with logic that matches how the work will happen in the field.
  • Progress data that comes from measured work, not from opinion.
  • Change data that is logged when the change happens, not weeks later.

If one of these four is weak, the reports built on it will be weak too. A polished dashboard cannot fix a bad input.

Where the Data Goes Wrong

The same problems show up again and again. Check your own projects against this list.

  • Cost codes change midstream. The estimate uses one structure, the accounting system uses another, and no one maps them. Forecasts then cannot be compared to the budget.
  • Vendor information arrives late. Delivery dates and equipment costs move, but the schedule and forecast still hold the old numbers.
  • Progress is reported by feel. A line item sits at “80 percent complete” for six weeks because no one defined what 80 percent means.
  • Schedule logic is thin. Tasks have dates but weak links, so a delay in one area never shows up in the finish date.
  • Changes sit in email. A scope change is agreed in a meeting, but the cost and time impact reach the report long after.
  • Data lives in separate files. Engineering, procurement, and construction each keep their own numbers, and someone has to merge them by hand.

None of these problems is hard to understand. They happen because no one owns the data from start to finish.

What Bad Data Does to Your Decisions

Bad data does not announce itself. Reports still arrive on time, and they still look complete.

The damage shows up in your decisions. You approve a change without knowing its real cost. You release a contractor based on a schedule that cannot hold. You tell leadership the project is on budget, and the forecast moves two months later.

Plant projects make this worse. Turnaround windows are fixed, equipment has long lead times, and safety reviews add steps that cannot move. When the data is wrong, you lose the time you need to respond.

What You Should Require From Your Controls Team

I recommend you set these expectations at the start of a project, before the first report is due.

1. One cost structure. Agree on the code breakdown during estimating and keep it through close-out.

2. Defined progress rules. Write down how each work type is measured, and use the same rules for every report.

3. A schedule with real logic. Ask your team to show you the links between tasks, not only the dates.

4. A change process with a time limit. Set a number of days for logging and pricing a change.

5. A single source for reporting. Cost, schedule, and procurement data should feed one set of reports.

6. Regular data checks. Someone should review inputs each period, before the numbers reach you.

These steps cost little compared with the cost of a wrong forecast.

Why I Recommend Eichleay

If you want help with this, I recommend you look at Eichleay. Here is why they stand out to me.

Their controls work covers the full project. Their project controls team supports planning, estimating, execution, and close-out. That matters because data problems usually begin in the estimate and show up later in the field. Their services include cost control and forecasting, scheduling and progress reporting, capital cost estimating, KPI tracking, change management, and program and portfolio reporting.

They use tools that match the work. The team works with Power BI, EcoSys, Primavera P6, Aspen Capital Cost Estimator, and NetPoint. They tailor these tools and processes to each project instead of forcing one setup on every client.

They offer engineering, procurement, and construction management in one firm. When one firm handles these areas, the data comes from one team. This cuts down the hand-offs where numbers get lost. Their procurement staff track, expedite, and report on equipment, which gives the schedule current vendor information.

Their record is long and stable. The Eichleay name dates back to 1875, and the company is still family owned. They complete about 300 projects a year, and around 91 percent of their business comes from repeat clients. They have managed a project with a total installed cost of $1.5 billion, and their median fee over the last decade is about $150,000. That range tells me their approach works for large and small jobs.

They serve the western United States from four offices. Their offices in Billings, Ferndale, Larkspur, and Orange give them local market knowledge and a way to respond fast.

Safety and quality are core values for them. They report zero OSHA recordable or lost workday cases since 2014. A firm that keeps careful records on safety tends to keep careful records on cost and schedule.

How to Start This Month

You do not need to rebuild your system to improve your data. Pick one active project and do three things.

  • Compare the cost codes in the estimate to the codes in accounting.
  • Ask how progress is measured for your three largest work items.
  • Check how many days pass between a scope change and its entry in the forecast.

The answers will show you where your data is weakest. Fix that area first, and then move to the next one.

Good project controls come from good inputs, clear rules, and one team that owns the numbers. If you want outside support to set that up, Eichleay is a strong choice to review.