Modular Skid Packages: Why Process Plants Build Off-Site

A modular skid package is a process system, pumps, vessels, piping, instrumentation, controls, built and tested on a structural steel base in a workshop, shipped as one or more units, and connected on site at a handful of flanged and electrical interfaces.

The alternative is stick-building: shipping loose components to site and assembling them in place. There are good reasons plants moved away from that, and a few situations where it is still right.

The case for building off-site

1. Workshop hours cost less than site hours

A welder in a workshop has overhead crane access, a bench, power, shelter, a stores counter and a fitter alongside. The same welder on a congested site has scaffold, a permit, a weather delay, and a walk to the tool store.

The productivity difference is large and well documented in industry practice. The precise multiplier varies, but nobody in the business argues about the direction.

2. Schedule runs in parallel

This is usually the biggest win and the least discussed. Skid fabrication happens while civil works are being poured. Stick-building cannot start until the site is ready.

On a project where the critical path runs through commissioning, moving eight weeks of mechanical assembly off the site critical path is worth considerably more than the fabrication saving.

3. Testing happens where testing is possible

A skid can be hydrotested, functionally tested, loop-checked and sometimes string-tested in the shop, with the vendor’s engineer standing next to it. Defects found there are cheap. The same defect found during site commissioning costs a mobilisation.

4. Quality is repeatable

Fixed positioning, proper fit-up, controlled welding conditions and an inspector who is permanently present produce better and more consistent work than site conditions allow. The documentation is also better, simply because the QC office is in the same building.

5. Site headcount falls

Fewer people on site means fewer permits, less congestion, lower accommodation and travel cost, and a smaller safety exposure. On remote or offshore projects this dominates every other consideration.

The case against, three situations where modular is wrong

Being honest about this matters more than the sales case.

1. When the transport cost exceeds the assembly saving

A skid is a large, awkward object. If the site is inland, the roads are poor, or the package breaks the shipping envelope, freight can eat the entire benefit. This is the main failure mode and it is arithmetic, not opinion, see out-of-gauge and breakbulk shipping.

Work out the freight before you commit to modular, not after.

2. When the design is not frozen

Modular front-loads engineering. Every interface, every tie-in point, every cable entry has to be decided before fabrication starts. A project where the process design is still moving will either delay the skid or build the wrong one.

Stick-building tolerates late change. Modular does not.

3. When the site already has the people

A plant with a permanent maintenance workforce and an idle workshop may genuinely be cheaper building in-house. The modular case rests on site labour being scarce, expensive or disruptive, where it is none of those, the case weakens.

Getting the modularisation split right

The engineering decision is where to cut. Some guidance from practice:

• Cut at flanges, not at welds. A site weld needs a qualified welder, a procedure, preheat capability and inspection, all things you moved off site in the first place.
• Keep instrumentation within a module. An instrument loop split across two modules becomes a site loop check.
• Minimise the number of interfaces: and make each one accessible. Ten well-placed flanges beat forty awkward ones.
• Terminate services at a single marshalling point per module, one cable entry, one earthing point, one utility header.
• Match the module envelope to the shipping tier: not to a convenient process boundary.
• Allow for site tolerance. Modules that must mate precisely need adjustment built in: slotted holes, spool pieces with machining allowance, or a designed shim pack.

What the structural base has to do

The skid frame carries three different load cases, and the governing one is often not the operating case:

1. Operating: equipment weight, thermal loads, nozzle loads, live loads
2. Lifting: the whole package on four lugs, possibly at a bad sling angle
3. Transport: longitudinal shock on braking, vertical shock over bad road, plus lashing loads

The transport case frequently governs. A skid designed only for its operating load can arrive twisted.

Frame design detail is in equipment skid design.

The interface document

The single most valuable deliverable on a modular project, and the one most often left until too late. It should specify, per module:

• Overall dimensions, dry and wet weight, centre of gravity
• Every mechanical interface: location, size, rating, orientation, face type
• Every electrical and instrument interface: terminal location, cable entry, earthing
• Foundation loads and bolt pattern
• Lifting arrangement: lug locations, rating, sling angle, spreader beam requirement
• Utility requirements at the boundary
• Clearances required for maintenance access

Get this agreed and frozen early. Almost every expensive modular surprise traces back to an interface nobody wrote down.

Testing scope: decide it at RFQ

What gets tested in the shop is a commercial decision, not a technical inevitability:

• Hydrotest of piping, standard, cheap, do it
• Functional test of the mechanical package, usually worth it
• Loop checks on instrumentation, worth it if the control system is in scope
• String test with the actual driver under load, expensive, needs shop utilities, and genuinely valuable on rotating equipment where site commissioning time is the constraint

Specify the scope at RFQ. Adding a string test after the package is built means re-mobilising.

A short project sequence

1. Freeze the process design
2. Decide the modularisation split, with freight costed for each option
3. Issue and freeze the interface document
4. Fabricate the structural base, stress relieve, machine mounting planes
5. Install equipment, pipe, wire
6. Test to the agreed scope, with your witness present
7. Preserve, pack, ship
8. Set, connect at the interfaces, commission

For what a skid order has to fix once you have chosen the modular route, see the steel skid fabrication guide.

Considering a modular route and unsure whether the freight kills it? That is the question worth answering first. Send the general arrangement and the destination and we will come back within 24 business hours with the structural scope and a freight tier.