By: Richard Keegan  |  From the Adjusting Matters Blog Series  | Part 4 – Linear BI Assumptions vs Real-World Stock Production

One of the most persistent challenges in Business Interruption (BI) claims is the assumption that business activity follows a neat, linear pattern. I was reminded of this recently while reading a book on big data, which suggested that humans naturally seek relationships between variables and tend to assume those relationships are linear.

This tendency is deeply embedded in the way many BI claims are presented, but in my experience, linear assumptions often lead to flawed strategies and under-reserving by Insurers. The real world is rarely that simple.

Here are a few examples from my own casework where non-linear relationships between production and output significantly influenced the claim outcome and where consistency of product, not just quantity, was critical.

Batch Production: Volume vs Time

Take a business that sells 1,200 tonnes of product annually. The instinctive assumption is 100 tonnes per month, and that production should be scaled accordingly. But nature and science don’t always cooperate.

Most production systems have a “sweet spot” for efficiency. Outside of this, diminishing returns set in. In many cases, products must be produced in batches, and throughput doesn’t scale linearly. The 80/20 rule often applies: 80% of yield may be achieved with 20% of the effort, but the final 20% takes disproportionately more time and resources.

Even in processes like cold rolling of steel where the hardness of the material increases linearly with reduction in cross-sectional area, the force required to achieve that change may not be linear.

So, the assumption that 1,200 tonnes annually equals 100 tonnes per month may be far from the truth.

Chemical Processing: Strength vs Time

In chemical production, the relationship between processing time and product strength often starts out linear but then flattens. For example, in one case involving a cyclotron irradiating a vial of liquid, the rate of radioactive absorption declines significantly after two hours as the liquid moves toward saturation.

Factoring in setup and clean-down times between runs, the most efficient production cycle isn’t the longest one, but the point where the marginal gain in strength per minute falls below the average gain over the full cycle. This kind of analysis is essential to understanding the true cost and timing of recovery.

Engineering: Mechanisation vs Manual Recovery

In another case, a UK manufacturer of aluminium car suspension sub-frames suffered a 24-hour shutdown after a forklift damaged overhead services. The business operated on a just-in-time basis, with parts collected daily for next-day use on a customer’s vehicle assembly line.

With no buffer stock and machinery already running 24/7, the company faced severe penalties. The solution? The introduction of multiple deliveries to the customer each day as stock was produced along with attempting a partial stock catch-up using manual fabrication at other plants at up to 16 times the cost of automated production.

One automated press, for example, could produce 16 sub-frames from a single sheet of aluminium with just one operator. Replacing that with manual labour was costly, but still more economical than the penalties for non-delivery. The final recovery plan included continued use of multiple daily deliveries before extending production during a planned shutdown to rebuild stock.

Agriculture & Food: Volume vs Quality vs Seasonality

In agriculture, seasonality and product quality are tightly linked. I once handled a claim involving a fire at a barn storing alfalfa (lucerne), grown for horse feed. The farmer harvested two crops per year, each with different nutritional profiles. To meet customer expectations, the two crops had to be blended meaning the loss of one crop affected the entire year’s output.

A similar issue arose with a seafood processor specialising in brown crabs. Crabs are only landed between April and September. Early in the season, they contain more brown meat, which is frozen and later blended with white meat to maintain a consistent product year-round.

In both cases, the loss of part of the annual production had a knock-on effect across the entire season, extending the indemnity period well beyond the immediate interruption.

Conclusion: The Cost of Linear Thinking

In all these examples, the temptation to reserve based on linear assumptions would have led to significant underestimation of the loss. As BI adjusters, one of our primary responsibilities is to provide early, accurate reserves for underwriters. To do that, we must go beyond the numbers and unpack the real relationships between resources, time, and output.

Understanding the qualitative features of stock its production complexity, seasonality, and the need for consistency is essential, because in the real-world recovery is rarely a straight line.