Why sheet-utilization percentage is not equal to nesting economic efficiency ART-141

What the utilization percentage actually measures

The usual formula looks simple:

Sheet utilization = part area / sheet area or available area × 100%.

Even here, however, the basis must be the same. One system may calculate full rectangular sheet area, another only the available area after margins and clamps. In some cases part area includes holes; in others it does not. A portion of the sheet can be reserved for a test cut, identification or a technological frame.

Two percentages can therefore be compared only when the following are identical:

• part geometry and revision;
• quantity of positions;
• sheet format and available area;
• material and thickness;
• rotation rules and rolling direction;
• gaps, margins and prohibited areas;
• treatment of holes, remnant and skeleton;
• algorithm completion criterion.

Without that, a “2% better nest” can simply result from different report settings.

Economic result has several components

For a management comparison, assess:

Total forecast cost = net material cost + machine time + gas and energy + preparation and completion labor + expected risk loss + logistics consequences.

This is not a universal accounting formula. It is a framework that makes costs beyond area visible. Modern CAM solutions themselves separate material utilization from production cost, and vendors describe optimization not only of material but also of movement, time, labor and data management.

| Criterion | What to check | Why yield gives no answer | |---|---|---| | Material | Net cost after reusable remnant and scrap | Equal unused area can have different liquidity | | Time | Cutting, piercing, idle moves, pauses | Dense nesting can sometimes complicate the path | | Risk | Part tipping, deformation, collisions | Percentage does not assess expected loss | | Labor | Programming, loading, sorting | Parts can be difficult to identify | | Lead time | Priorities and order compatibility | The best shared nest can wait for future parts | | Remnant | Shape, ID, circulation | A large but complex contour may never be used |

Example of two nests

Scenario A has 90% sheet utilization. Parts are tightly mixed, there are many small internal elements, estimated time is longer, and five orders require manual identification after cutting. The skeleton breaks into complex fragments.

Scenario B has 87%. Parts are grouped by order, the path is shorter, sorting is easier, and a rectangle remains at the side which warehouse accepts as a controlled remnant. If the cost of the additional 3% of new metal is less than saved time and labor plus expected remnant value, scenario B is more profitable.

This is not a rule that “lower yield is better.” Actual components must be compared. In a series of simple identical parts, material percentage can be the dominant factor. In mixed short orders, time, traceability and lead time become decisive.

Cutting time and number of events

Contour length alone is not equal to cycle time. Time is affected by number of pierces, approaches, transitions, speed changes, small-contour characteristics, sequence and technological pauses. Denser nesting can reduce idle movement, but may sometimes need a more complex sequence to keep the sheet stable.

Hypertherm describes direction and sequence control, simulation, common-line, collision avoidance and skeleton cut-up among ProNest functions. SigmaNEST states that nesting for complex machines considers clamps, movement, bevels and secondary processes. This does not prove that a particular automatic nest is optimal; it confirms that the real task has multiple criteria.

Compare estimated NC-program time using the same technological basis, then verify it against fact. Detailed plan/fact calibration is the topic of ART-142.

Risk of unstable sheet and parts

When a significant part of a sheet has been cut out, skeleton stiffness changes. Small parts can shift or tilt; large contours can affect support. Risk depends on machine, table, material, thickness, geometry, sequence and verified strategies.

Economic risk can be represented as:

Probability of an undesired event × expected cost of its consequences.

This does not mean an operator should independently change technology. Manufacturer documentation, approved modes and technologist assessment apply to a particular machine. For calculation it is sufficient to see that aggressive compaction is not free if it historically raises stops or rejects.

Reusable remnant versus an attractive percentage

Two nests can have equal unused area. The first leaves narrow strips and pockets between parts; the second leaves one rectangular remnant. For the next order, the latter is often more useful if it can be identified, stored and issued from warehouse.

SigmaNEST’s official description connects remnant nesting with inventory priority, status and value. TRUMPF describes separate remnant management in the production loop. The practical conclusion is to add expected reuse suitability to nesting criteria, not merely area.

Sometimes it makes sense deliberately to end a nest on a straight line even if the algorithm could insert several more small parts into the future remnant. The decision depends on actual product mix: if such remnants are never used, their imagined value is zero or scrap value.

Sorting and traceability

A mixed nest can raise yield, but after cutting a person must sort parts without error by order, revision and subsequent operation. Where positions are similar, mis-sorting risk grows. Additional marking can help, but it also has time, quality rules and technological limits.

Record when comparing scenarios:

• expected sorting minutes;
• number of orders and different positions on a sheet;
• share of parts difficult to distinguish visually;
• identification method;
• actual kitting errors;
• repeat movements and waits for the next operation.

If metal saving is a nominal UAH 300 but additional sorting and correction cost UAH 600, the “better” yield worsens the result.

Order lead time as an economic criterion

An algorithm can wait for an ideal pool of parts with the same material and thickness. A customer order, however, has a due date. Saving half a sheet does not justify delay, penalty, stoppage of the next operation or loss of trust.

Set an allowable waiting window. Nesting for every order group is optimized inside that window, not indefinitely. Urgent work can have separate economics described in ART-140. Grouping without breaking due dates is covered by ART-133.

How to create an integrated indicator without false precision

It is not necessary to reduce everything to one score. A comparison card is often more reliable:

| Indicator | Scenario A | Scenario B | Data source | |---|---:|---:|---| | Net material cost | | | Warehouse + CAM | | Estimated time | | | CAM | | Pierces / technological events | | | NC report | | Programming | | | Labor record | | Sorting | | | Standard / fact | | Reusable remnant | | | Warehouse register | | Risk, UAH | | | Event history | | Due-date fulfilment | yes/no | yes/no | Plan |

First exclude scenarios that do not pass safety, quality or lead time. Among admissible scenarios, compare total forecast cost. Yield remains important, but cannot override mandatory limits.

How to test the decision experimentally

1. Select 10–20 typical order sets. 2. Create at least two admissible nests for each using the same basis. 3. Retain CAM version, settings and reports. 4. Before launch, record material, time, sorting and risks. 5. Run programs only within approved technology. 6. Record actual cycle, stops, labor, yield and remnant. 7. Convert differences to money using internal rates. 8. Update nesting rules from repeatable results.

One successful sheet is not sufficient statistics. Geometry, thickness and order mix change the balance of criteria. Retain results as plan/fact pairs so the next decision uses not only CAM forecast but the history of similar product mix.

Admissibility first, then economics

Not every option needs immediate conversion into money. If a nest fails technological verification, breaks a mandatory due date or does not provide traceability, exclude it before calculating savings. Otherwise very low material cost can conceal a scenario production is not entitled to execute.

In practice the decision has two levels:

1. Admissibility filter. Check approved technological limits, part revisions, material, ability to meet due date, identification and acceptance rules. 2. Comparison of admissible options. Calculate material, machine time, preparation, sorting, remnant, expected loss and flow effect.

This prevents a false “trade-off” where an unsafe or late nest is justified by a few material percent. It also simplifies the technologist’s work: no detailed costing is needed for an option that already violates a mandatory condition.

Different order classes need different priorities

One comparison card can be used for all work, but criterion order need not be identical. For serial parts with stable demand, reusable remnant and cycle repeatability can matter more than starting a particular sheet quickly. For a short urgent order, lead time and minimum preparation can outweigh a small difference in metal utilization.

Define classes beforehand:

• serial recurring parts;
• mixed short orders;
• parts with strict traceability;
• work with high material cost;
• urgent orders with a confirmed due date;
• orders where ease of post-cutting kitting is critical.

For every class, record decision logic rather than arbitrary weights. For example, first due date and traceability, then total forecast cost; or first technological stability, then material result and cycle duration. If weights are used, validate them on actual orders and do not hide them behind one “efficiency index.”

The cost of information also has a limit

Comparing dozens of options is not always useful. Extra technologist time is justified when the potential difference in material, duration or risk is material to the batch. For a simple inexpensive sheet, hours spent seeking the theoretical best nest can cost more than possible saving.

Set a stopping condition before optimization:

• maximum time for alternative nests;
• minimum expected difference worth checking;
• orders requiring mandatory comparison;
• approval level for deviating from a standard scenario.

If two admissible variants yield close total cost within data uncertainty, the simpler, more reproducible one can be better. Apparent precision to a kopeck does not improve a decision if time, scrap price or sorting effort are approximate.

How to read plan/fact differences

After execution, determine not only which option “won” but why. If material matched forecast but sorting took longer, the problem may be mixing similar parts. If machine cycle differs, separately check CAM-time accuracy. If a planned remnant did not enter warehouse, nesting economics were overstated even when geometric percentage was correct.

Recurring causes become preparation rules: when to group parts, preserve a rectangular remnant, limit order mixing and spend time on an alternative scenario. This decision base—not a record percentage on one sheet—creates a durable economic effect.

The decision owner must also be defined. The CAM technologist prepares admissible options and explains technological consequences; the planner sees due dates and queue; warehouse confirms remnant suitability; the cell manager approves exceptions. When each person measures only their own indicator, a locally “better” nest easily worsens the total result.

Typical mistakes

Rewarding a technologist only for yield. It encourages sacrificing time, due date or usable remnant shape.

Comparing percentages on different bases. Use identical parts, sheet, margins and constraints.

Treating a vendor claim as a guarantee. Average-saving data are a marketing reference, not your product-mix result.

Ignoring downstream. Cutting can be quick while the cell waits for sorting, bending or kitting.

Assessing risk in words. Even a simple stop-and-reject log is better than arguing that a nest “looks unsafe.”

Selection checklist

• Is comparison made on an identical basis?
• Do all scenarios pass technology and safety limits?
• Are machine time and event count included?
• Are programming and sorting labor assessed?
• Does remnant actually return to stock?
• Is the order fulfilled on time?
• Are risks based on facts?
• Is plan/fact collected after cutting?
• Does the optimization criterion fit today’s cell objective?

Conclusion

Sheet utilization is a necessary technical indicator but a weak single business criterion. An economically efficient nest minimizes not empty area but total cost of an admissible production scenario, including time, labor, risk, remnant and due date.

The most practical approach is not to seek a universal “perfect” nest, but compare two or three scenarios using one card and calibrate CAM with actual data. Higher yield then remains an advantage only where it genuinely creates money, not a beautiful report percentage.

Need help selecting equipment?