Collected tonnes are not saleable tonnes

These four terms are used interchangeably in project discussions and mean different things. Keeping them apart is what prevents a business case from overstating production.

Mechanical capacity is the maximum the plant can process per hour if it runs at its design rate on a suitable input. It is a property of the equipment, established by the supplier.

Utilisation is the share of available hours the plant actually operates. Shifts, maintenance, cleaning, changeovers, breakdowns and the availability of input all determine it. A plant designed for continuous operation that receives material five days a week does not run at design utilisation.

Input throughput is what actually enters the plant: the mechanical capacity multiplied by the utilisation, reduced by the effects of difficult input — a heavily contaminated or light bale feeds more slowly and jams more often than a clean, dense one.

Sold output is input throughput multiplied by the yield of saleable product, which is where the losses described above apply. It is the only figure that generates revenue, and it is typically substantially below the mechanical capacity the plan began with.

The compounding is the point. A plan that starts from mechanical capacity and applies a single loss factor understates every stage of the reduction, and the error grows with each stage it skips.

The discipline that keeps the analysis honest is a mass balance that reconciles to zero. Every kilogram entering is accounted for as either saleable output, a named loss stream, or stock movement. The table below is an illustrative balance with assumed figures chosen to make the structure visible; it is not a plant case, and the numbers are not drawn from any real facility.

The inputs are receipts plus opening stock. The outputs are saleable grades, the loss streams, and closing stock. If the columns do not reconcile, something has been omitted or double counted, and the exercise of finding it is the exercise that makes the plan reliable.

The reason grades matter is that they sell at different prices. A tonne of clean, single-polymer material commands a different price from a tonne of mixed or contaminated material, and the difference is frequently larger than the difference in volume.

This changes the incentive structure of the project. Reducing contamination improves both the volume of saleable output and its grade, and the effect on revenue is the sum of the two. Conversely, an assumption that all output sells at a single blended price conceals the sensitivity of the project to input quality — which is precisely the variable the project can influence through its sourcing.

A plan should therefore carry, for each grade, an assumed price and the specification that grade must meet. Where an off-taker has agreed a specification, that is the evidence; where the specification is assumed, it should be labelled as an assumption, and the sensitivity of the case to it tested.

The quality of the input and the specification of the output are therefore linked directly to revenue, and the mass balance is what connects them. A project that cannot reconcile its mass cannot state its grades; a project that cannot state its grades cannot state its revenue.

Building the reconciliation first changes what the project plans for. It directs attention to input quality as a commercial variable rather than as a nuisance, because contamination reduces both volume and grade. It makes the sorting and reprocessing steps visible as loss centres with costs attached, rather than as fixed characteristics of the process.

It also changes how capacity is described. A plan that states mechanical capacity as though it were output is overstating the project by however much the loss chain removes, which in the illustration above is roughly a fifth of receipts even before utilisation is considered. The honest description of a recycling project's size is its saleable output, not the tonnage it can accept.

The final discipline is to keep the assumed figures labelled. A mass balance built on assumptions is a legitimate planning tool, and it becomes misleading the moment its figures are presented as results. The plan should say what is assumed, what has been tested, and what will be measured once the plant operates — and the same balance should be rebuilt from measured data at the first opportunity.