To compare loss, the measurement must be unified first

Mass loss during roasting is usually calculated from the difference between pre- and post-roast weight. The formula sounds simple, but the result can only be compared if the two weighing points are defined by the same rule. Weighing right after leaving the machine and weighing after a period of cooling don't necessarily give the same result.

Research on heat and mass transfer in batch roasting shows mass can be tracked online alongside temperature and humidity, describing the process's kinetics. This emphasizes that the timing of measurement is part of the measurement itself, not a minor detail.

The same formula doesn't mean the same data

If pre-roast mass is called M1 and post-roast mass M2, the mass-loss ratio can be calculated from the difference relative to the initial mass. But for two batches to be comparable, M1 needs to be determined the same way, M2 needs to be weighed at the same state, and weighing conditions need to be equivalent.

If one batch is weighed while still hot and another after fully cooling, the difference could reflect the entire post-roast process too. So standardizing the weighing point matters just as much as the calculation formula.

Raw material and heat change the loss level

Initial moisture is a clear variable. Beans with different water content will have different amounts of volatile material during roasting. Temperature and time also change the kinetics of mass loss.

Vargas-Elías and colleagues studied Arabica roasted in a gas-heated system and found the rate of mass loss depends on thermal conditions. This research is useful for understanding the mechanism but shouldn't be used to apply a fixed loss ratio to every type of coffee.

Charge weight needs to be controlled

A small batch and a large batch on the same machine can have different heat-transfer dynamics. If charge weight changes without being recorded, a difference in loss could be mistaken for a difference in roast level.

When standardizing, you should record charge weight, raw-material type, moisture if available, initial temperature, equipment configuration, roasting time, endpoint, and post-roast weight. This data creates the necessary context for explaining a loss figure.

The scale is also part of the measurement system

The accuracy and repeatability of the scale need to suit the mass being measured. Different scales can produce small but meaningful discrepancies when the goal is comparing many batches. Scale placement, mounting surface, and environmental conditions also need to stay stable.

If a facility wants to use loss ratio as a long-term tracking indicator, it should periodically check the scale's condition and standardize how samples are placed. A neat number on a spreadsheet doesn't have much value if the measurement behind it isn't stable.

Loss is a control indicator, not the sole measure of quality

Two batches can have a similar mass-loss ratio yet differ in thermal history, color, aroma, or sensory quality. Bottazzi and colleagues modeled the roasting process based on heat and mass transfer, showing many system variables jointly contribute to the final result.

So the loss ratio should be read alongside temperature, time, and relevant quality indicators. It's especially useful for detecting a batch deviating from its own history.

When there's a deviation, the data needs to help find the cause

A good standardization system doesn't just record the loss ratio — it also sets a rule for what to do when a result exceeds the reference range. You can recheck the weighing, raw-material moisture, charge weight, temperature, time, and cooling conditions before concluding there's a product problem.

If the variables are fully recorded, investigation moves from a sign to a cause rather than adjusting the machine based on guesswork. This is the biggest practical value of standardization: turning a loss number into a trace that can be tracked across batches.

When an unusual loss is found in a batch, pre- and post-roast data should be checked together. A single deviated ratio isn't enough to conclude the roaster has a problem; you need to look at charge weight, raw material, temperature, and the weighing method as well.

After roasting ends, beans continue cooling and releasing gas. If one batch is weighed right after cooling while another is weighed after a longer wait, the result can differ even though the roasting process was the same. So the timing and post-roast weighing state need to be clearly specified in the procedure.

Standardization needs to include post-roast state

When standardizing across batches, you should record both the absolute value and the percentage. The absolute value shows how much mass was lost, while the percentage helps compare batches with different charge weights. The two views complement each other and help catch cases where data looks normal by percentage but abnormal in actual mass.

Another point worth agreeing on is the cooling method before weighing. If one batch is cooled one way and another differently, the gas released and thermal state of the sample can differ. The procedure should clearly define when a sample is considered ready to weigh and keep that handling consistent across measurements. Once post-roast conditions are stable, the loss-ratio deviation can better reflect actual differences in the roasting process.

References

• Hernández et al. (2007). Journal of Food Engineering, 78, 1141–1148. https://doi.org/10.1016/j.jfoodeng.2005.12.041
• Vargas-Elías et al. (2016). Engenharia Agrícola, 36, 300–308. https://doi.org/10.1590/1809-4430-Eng.Agric.v36n2p300-308/2016
• Bottazzi et al. (2012). Journal of Food Engineering, 112, 243–252. https://doi.org/10.1016/j.jfoodeng.2012.04.009