Cutting stock optimiser

How the Cutting Optimiser Works

The OptimalCutter engine evaluates your required cuts, available stock lengths and optional constraints, then runs several optimisation strategies in parallel. Each method generates its own set of cutting patterns, and the system automatically selects the option with the lowest waste. This gives you the best output every time without needing to understand the algorithms behind it.

1. Required Cuts

You provide the lengths and quantities of the pieces you need. These define the total demand. Each cut can also include an optional label to help identify where the piece will be used.

2. Stock Lengths

You also provide the available stock lengths and their quantities. These can be timber, aluminium, steel, plastic or any other linear material. Stock lengths can optionally be marked as “priority” to encourage the optimiser to use them first.

3. Optional Constraints

• Trim length: Ensures a minimum trim allowance at each end of a bar.
• Kerf thickness: The amount of material removed during each cut due to the blade width.
• Max cuts per bar: Limits the number of cuts on one bar if tooling or workflow requires it.
• Over-production: Allows the optimiser to produce extra pieces if it significantly reduces waste.

These settings modify every optimisation method equally, ensuring fairness and consistency between algorithms.

4. Multiple Optimisation Methods

To achieve the best result, OptimalCutter does not rely on a single algorithm. Instead, it runs several cutting-stock optimisation methods simultaneously:

• First Fit Decreasing (FFD): A fast, reliable heuristic that sorts cuts by length and fills bars in order.
• Best Fit Decreasing (BFD): Chooses the bar with the smallest remaining space that can accommodate the next cut.
• Greedy Pattern Builder: Generates locally optimal patterns by repeatedly selecting the longest remaining cut that fits.
• Bin-Packing Variant: Treats stock lengths as bins and attempts to pack required cuts efficiently using classical bin-packing heuristics.
• Metaheuristic Hybrid: Uses a guided trial-and-improvement approach inspired by genetic algorithms. It mixes mutation, crossover and local search ideas to discover patterns that simple heuristics may miss.
• AI-Assisted Variant: Learns from previous optimisation attempts to favour cut combinations and stock selections that historically lead to lower waste. This helps the optimiser adapt over time.

5. Automatic Best-Result Selection

Once all methods have generated their cutting schedules, the engine compares:

• Total waste
• Number of cuts
• Use of priority stock
• Conformance with constraints (trim, kerf, max cuts, etc.)

The solution with the lowest total waste is selected automatically and returned as the final output. This means you always get the most efficient option available without needing to choose a method manually.

6. Transparent Rules

All methods use exactly the same input data and constraints. None of the algorithms are allowed to “cheat” by ignoring kerf, trim or stock limits. This ensures consistent, predictable behaviour and gives you confidence that the final schedule is based on fair comparison.

7. Final Output

The results include:

• Bar-by-bar cutting patterns
• Colour-coded segments to show each cut
• Waste amounts per bar
• Total waste, trim waste and kerf waste
• Number of cuts
• CSV export for use in workshops, manufacturing or documentation

The goal is simple: help you cut material efficiently, save money on off-cuts and improve workflow with clear, optimised cutting schedules.