Design for Manufacture Cost Model for PCB

A design for manufacture cost model is a small calculation that connects layout choices to the price of a board. It does not need to be precise to be useful; it needs to be fast, consistent and visible at the moment the choice is made. A designer who can see that two more layers add a third to the unit cost, or that moving the outline by a few millimetres recovers a whole column of panels, makes better decisions without any additional skill.

What the Model Is For

The model exists to make the cost consequence of a decision visible while the decision is still open. Once a stackup is released, the material, the layer count and the panel size are fixed, and the remaining savings are small. A model that is accurate to within ten or twenty percent and available at the design review is worth far more than an accounting figure that arrives three weeks after the layout is frozen.

The second purpose is comparison. Two routes to the same function rarely cost the same, and the difference is usually in the manufacturing cost rather than in the component list. A six layer board with a tight outline may cost more than an eight layer board that fits a standard panel, and the only way to see that is to run both against the same model with the same assumptions about volume and yield.

The Cost Elements

A workable model has five blocks: material, fabrication process, assembly, test, and the overhead that is applied on top. Material includes the laminate, the copper, the solder mask, the surface finish and the prepreg, and it is the block that varies most with the stackup. Process cost is the time each step takes multiplied by the rate of the equipment, and it grows with layer count, aspect ratio, hole count and finish complexity.

Assembly and test are usually the largest blocks in a populated assembly and the smallest in a bare board, which is why the same model has to be able to switch between them. Overhead is applied as a rate rather than entered item by item, because the purpose of the model is comparison and a consistent overhead treatment compares routes fairly. Choosing the layer count is the single design decision that moves most of these blocks at once.

Fixed and Variable Costs

Fixed costs, such as tooling, artwork, test fixtures and the first article inspection, do not change with the number of boards built. Variable costs, such as material, drilling time, plating time and test time, repeat with every unit. Separating them is what allows a sensible price to be set for a small order without charging the same unit cost that applies to a large one.

The fixed block is what makes small orders expensive, and it explains why a customer who asks for fifty boards pays several times the unit cost of a customer who asks for five thousand. A model that blends everything into a single rate hides this, and hides it in the direction that causes the most arguments, because the small order then looks unprofitable for reasons nobody can point to.

Panel Utilisation

Panel utilisation is the ratio of board area to panel area after the rails, the tooling strip and the gaps are removed. It is the most underrated term in the model, because a small change in outline dimensions can move it several percent and the effect falls straight through to the price. Rotating a board on the panel, changing the gap, or moving the tooling holes into the rail can all recover area at no cost.

In practice the utilisation is calculated for each candidate layout and the cost difference is reported as a percentage rather than as a lecture. A designer who sees that a two millimetre trim lifts utilisation from seventy eight to eighty four percent can usually find the trim, while the same designer confronted with a general instruction to use material efficiently has nothing concrete to act on.

Where Design Decisions Bite

Layer count is the largest single lever, followed by aspect ratio, hole count, minimum trace and space, and surface finish. Each of those choices adds process steps or reduces tolerance, and process steps cost money in proportion to the number of panels that pass through them. A finish change alone can alter the price by a noticeable fraction, and it is usually agreed at the last minute without any cost discussion.

Tolerance is the second lever that designers underestimate. Calling out a tighter tolerance than the function requires forces extra inspection, lowers yield and occasionally forces a different process, and the cost appears in the yield rather than in the quotation. Outline tolerance rules are a good example, because the drawing value is often copied from an old project rather than derived from what the assembly actually needs.

Using the Model in Review

The model earns its place when it is used in a design review as a live tool. The designer proposes a change, the cost consequence is estimated in a minute, and the decision is recorded together with the assumption. That record is what makes the next review faster, because the same questions do not have to be argued again from first principles with a different set of numbers.

Reviews should also feed the model. If a quotation later shows that a particular stackup or finish costs more than the estimate, the rate behind the estimate should be corrected, and the correction should be written down with the fabrication notes for the next project. A model that is never corrected drifts away from the shop’s real costs and is quietly abandoned, which is the usual fate of manufacturing cost reduction tools held by one person.

Practical Rules

Keep the model simple enough to run in a meeting, and use the same assumptions for every option being compared. Report the difference as a percentage of unit cost so that the size of the effect is obvious, and always show the fixed block separately so that volume effects are visible rather than buried.

Review the rates once a quarter against real quotations, and correct them without ceremony. The value of the model comes from the habit of asking the cost question at the point of design rather than after release, and a habit survives only if the tool is fast, consistent and trusted by the people who use it. A model that nobody trusts is worse than no model, because it produces confident answers that the review then has to argue away.

FAQ

How accurate does a cost model need to be? Accurate enough to rank options correctly. A consistent estimate within ten or twenty percent that is available early is more useful than a precise figure that arrives after release.

Which design choice affects cost most? Layer count, because it changes material and nearly every process step at once. Outline size and panel utilisation are the next largest.

Why separate fixed and variable costs? Because the fixed block is what makes small orders expensive. Separating it allows a fair price for both small and large quantities.