How Good Is the Optimal Algorithm? A Real Example

We took a real cut list that trips up ordinary algorithms and ran it through the optimal solver. The result isn't just a “good” solution — it's the mathematically proven optimum.

▶ Open the calculator

The problem

Stock: 10 × 6000 mm. Parts: 10 × 2900, 5 × 1900, 10 × 1180, 15 × 589 and 4 × 170 mm — 59,815 mm of parts against 60,000 mm of stock with a 1.5 mm kerf. A tight instance: only 134 mm of slack after kerfs.

The results

Plain greedy nesting (FFD) leaves 1,178 mm uncut. A popular alternative tool left 589 mm on the same input. CuttingOpt's optimal algorithm finds an arrangement with just 340 mm uncut — a 99.1% yield — and an exhaustive search proved no better solution exists for these lengths.

Where the difference comes from

Single-pass algorithms place parts one by one and never see combinations like 2 × 2900 + 170 on one bar. The optimal solver looks at all bars at once and coordinates the patterns between them.

Frequently asked questions

Does it always find the optimum?

For problems up to a few hundred parts — yes, usually in under a second. On very large inputs it returns the best solution found within its time budget.

Can I reproduce this test?

Yes — enter the numbers above into the calculator and pick the “Optimal” algorithm.

▶ Open the calculator

More on cutting optimization