Business case

Business Case: Create a Robust French Fries Process

This business case starts from zero.

Imagine a small food business wants to create better French fries, or frites, without buying new equipment or stopping service for weeks of testing.

The owner does not need a scientific laboratory. The owner needs a clear way to improve daily work: make fries that customers like, keep service moving, reduce waste, and create a process the team can repeat.

Taguchi helps turn that business problem into a practical experiment.

Business situation

The shop already sells fries, but quality changes too much from batch to batch.

Some portions are crisp and golden. Others are soft, too oily, pale, overcooked, or inconsistent. The team has opinions about what helps, but the process is not written down clearly enough to train new staff or explain why one batch worked better than another.

The owner wants the “perfect” fries process for this business, not in theory. That means the process must fit normal ingredients, normal staff, one fryer, real lunch pressure, and customer expectations.

The business question becomes:

Which process settings give consistently good French fries while respecting our time, equipment, cost, and staffing limitations?

Business objectives

The first step is to define what better means. For this shop, the business objectives are:

• Crisp outside without becoming hard.
• Soft inside without becoming raw or wet.
• Golden color that looks appetizing.
• Balanced salt and taste.
• Fast enough service during lunch.
• Lower waste from rejected or remade portions.
• A method that staff can repeat without constant supervision.

These objectives are connected. A very long fry time may improve crispness, but it may hurt service speed. A high oil temperature may improve color, but it may increase burning risk. A business decision needs balance.

Business limitations

The limitations make the problem real:

• The shop has one fryer.
• The team cannot run dozens of test batches.
• Potatoes vary by supplier and season.
• Oil temperature drops when the basket is full.
• Customers will not wait for a slow premium process.
• Staff need simple instructions during busy service.
• Ingredient cost and oil life matter.

These limitations should not be hidden. They are part of the process design. A robust process is useful because it works inside these limits.

Taguchi framing

Taguchi connects objectives and limitations by asking the business to choose controllable factors, practical levels, and one measurable response to optimize.

The owner lists possible factors:

• Potato type
• Cut thickness
• Soak time
• Oil temperature
• Fry time
• Basket load
• Resting or draining time
• Salt timing

Then the owner chooses only the factors that can realistically be tested first. For example:

The owner may care about several measurements:

• Crispness score
• Inside texture score
• Taste score
• Time to serve
• Oiliness
• Waste

TaguchiLab currently has one response input per combination, so the business must turn those measurements into one decision number.

There are two practical ways to do that:

• Use a primary response if one result matters most. For example, enter only a crispness score if the first experiment is about crispness stability.
• Use a composite quality score if the decision must balance several objectives. For example, score each batch from 1 to 10 by combining crispness, inside texture, taste, speed, oiliness, and waste using the same rule every time.

The extra measurements still matter. They can be written in a notebook or spreadsheet, then used to calculate the single value entered in TaguchiLab. The important part is consistency: every combination must use the same scoring rule.

The owner also chooses the response direction:

• Larger is better when a higher value is better, such as crispness score, taste score, customer rating, or a composite quality score.
• Smaller is better when a lower value is better, such as time to serve, oiliness, waste, defects, complaints, or cost.

Why not test one thing at a time?

Changing one thing at a time can feel simple, but it becomes slow when many factors might matter. It can also miss combinations. A medium cut with a high oil temperature may behave differently from a thin cut with the same temperature.

Taguchi gives the owner a smaller, structured test plan. The business can learn from several factors without testing every possible combination.

TaguchiLab helps keep that plan organized: factors, levels, runs, and the single response value stay together so the owner can compare results instead of relying on memory.

Business decision

After the test batches, the owner does not simply choose the crispiest batch. The decision must match the whole business.

For example:

• Batch A is very crisp but takes too long.
• Batch B is fast but too soft.
• Batch C is crisp enough, soft inside, consistent, and fast enough for lunch service.

Batch C may be the better business decision because it balances quality, speed, waste, and repeatability.

This is where Taguchi connects to a robust process. The chosen settings should work well under normal variation, not only in one lucky test.

Operating standard

The final result is a simple operating standard for the team.

Example:

• Use medium cut fries.
• Soak for 20 minutes when prep time allows.
• Fry at 175 C.
• Fry for 4 minutes.
• Drain for 45 seconds.
• Salt immediately after draining.
• Keep basket load below the amount that drops oil temperature too much.

The numbers above are not universal. They are an example of what the business would confirm with its own data.

What improved

The business now has:

• A clearer definition of good fries.
• A documented process staff can repeat.
• A way to explain tradeoffs between quality and speed.
• Evidence for the chosen settings.
• A practical path for future improvements.

The point is not that Taguchi magically creates perfect fries. The point is that it helps a business learn from a small set of well-planned tests and turn that learning into a robust process.