What Does a Machine Vision Feasibility Study Cost, and Is It Worth It?

Machine vision projects fail more often at the buying stage than at the engineering stage. A company decides vision will solve their inspection problem. They buy a system on that assumption. The first time anyone points a camera at the real part is after the money has been spent. Sometimes it works. Sometimes the part is more reflective than anyone allowed for, or the defect has almost no contrast, or the line runs faster than the setup can handle. By then it is a difficult conversation rather than a cheap fix.

A feasibility study is how you avoid that outcome. You bring your actual parts to a lab. They are tested with real cameras, lenses and lighting. You find out whether vision can do the job before you commit to a system. This article explains what that study involves, what it costs, what you get at the end and why testing first is the cheapest insurance you can buy on a vision project. It is written from the perspective of the team that runs those tests at Clearview every week.

What the Insights Lab actually is

The Insights Lab is our dedicated test facility. It is not a showroom with tidy demos of parts that were always going to pass. It is a working lab where we put your real products in front of real vision hardware and find out what happens.

You send us samples, ideally a spread that covers the good, the bad and the borderline. We image them across different camera resolutions, lens choices, lighting types and angles. We work out which combination brings out the feature or defect you care about. Some applications are solved in an afternoon. Others take longer. The hard part of vision is very often the lighting. Getting that right is a matter of testing rather than guesswork. The point is that we are proving the approach on your parts, not on a spec sheet.

Why testing first matters so much

Machine vision lives or dies on the image. If the image clearly shows the thing you need to inspect, the rest is tractable. If it does not, no amount of clever software will reliably rescue it. That single truth is why a feasibility study matters more in vision than in almost any other kind of automation.

The trouble is that whether you can get that image is not always obvious in advance. A part that looks simple can be a nightmare of glare and reflection. A defect that is glaring to the human eye can be nearly invisible to a camera. One that seems subtle can light up instantly under the right wavelength. Experience narrows the options quickly. Testing is what turns a confident hunch into evidence. Skipping that step is where projects quietly go wrong.

What you get at the end

A feasibility study is only worth doing if it leaves you with something you can act on. Here is what you get from an Insights Lab study.

The written report is the part that earns its keep inside your business. Most vision projects need sign-off from someone who controls a budget. A report showing tested results on your own parts, with an indicative system cost, is a far stronger case than an engineer saying they think it will work. It turns a request for money into an evidence-based proposal.

What a feasibility study costs

This is the question everyone wants answered. Here is the plain version. For a great many enquiries, the initial feasibility study is complimentary. We test your parts, form a view and share what we find as part of understanding whether we can help. We would far rather spend that time upfront than sell you a system that turns out to be wrong for the job.

Some studies need significant engineering time. More complex trials, deep learning proof of concept work, extended testing across many part variants. Those may carry a charge. When that is the case we agree it with you before we start. The cost is never a surprise. The fee, where there is one, reflects the engineering hours that go into proving out a harder application. It is modest set against the cost of getting a full system wrong.

It is worth putting that in proportion. A custom machine vision system is a serious investment, often running from £15,000 to £100,000 or more depending on scope. Against a number like that, a feasibility study is a small fraction of the project cost. It is spent at the exact point where it removes the most risk. That is the whole economic case for testing first.

The return on testing before you commit

The return on a feasibility study comes in three forms. Each one is worth more than the study costs.

It stops you buying the wrong system. The most expensive vision system is the one that does not work. If testing shows a given approach will not hit the standard you need, you have saved the entire cost of finding that out the hard way, on the line, in production.

It gets the specification right. Testing does not just answer yes or no. It shows which camera, which lighting and which approach actually work. The system you go on to buy is built on evidence rather than assumption. That is the difference between a system that works first time and one that needs expensive rework.

It builds the business case. The report gives you the evidence and the indicative cost to take internally. That shortens the path to approval. A tested proposal moves through a business far faster than a hopeful one.

There is a fourth return that is harder to put a number on. Sometimes the straight answer is that vision is not the right tool for a particular job today, or that the technology is not quite there for your specific challenge. Finding that out in a lab, for little or nothing, is enormously more valuable than finding it out after a system has been designed, built and installed. We would always rather tell you that early.

When a feasibility study is the right next step

A study is worth doing whenever the imaging is not a certainty. That is most of the time. It is especially worth it when the parts are shiny, dark, transparent or awkwardly shaped, when the defect is subtle or variable, when the line runs fast, or when you are going to have to justify the spend to someone else. If any of those describe your application, testing first is the sensible move.

There are cases where you can reasonably skip it. If the application is a close match to something proven many times over, a straightforward barcode read on a flat label, for instance, the risk is low and the path is clear. Even then, a quick assessment costs you nothing and confirms the approach. When you are unsure, testing is always the lower-risk choice.

Test first, then invest

A feasibility study is the step that turns machine vision from a gamble into a calculated decision. For most enquiries it costs nothing. Where it does carry a charge, that charge is small against the system it protects. It is agreed with you upfront. What you get back is evidence, a clear verdict, a realistic budget and a report you can build a business case on. Set against the cost of a system that does not perform, it is the best value money you will spend on the whole project.

If you have an inspection challenge and you want to know whether vision can solve it, the fastest way to find out is to let us test your parts. Bring us the good, the bad and the borderline. We will show you what is possible before you commit a penny to a system.

Book your Insights Lab session: test your parts with us

Get in touch: info@clearview-imaging.com | +44 (0)1844 217270

Related: Insights Test Lab   |   Pricing   |   The True Cost of Machine Vision   |   Machine Vision for SMEs