Every answer and its explanation appears here once you have finished the path. Each one then links to the matching glossary entry, where the concept is set out in full with its worked example.
1. In what way is a construction site the opposite of what property insurance seeks?
It deliberately gathers a growing value at a single point, when insurance seeks dispersion, and that explains almost all underwriting in this line
The other three answers describe real difficulties of a site and do not explain underwriting; this one explains almost all of it, because a property insurer lives on dispersion and a site undoes it by design. That is where the place of probable maximum loss comes from, the attention to the schedule, and the fact that capacity is negotiated before price. Retaining the absence of loss history leads toward a rating argument, when the subject is capacity.
Glossary entry →2. Which figure drives the capacity and price of a construction policy, and why that one?
The probable maximum loss, which approaches the present value for want of compartmentation, roofing and detection
On an operating industrial site, probable maximum loss is far below total value because firewalls, roofs and detectors stop a loss; on a construction site, none of that exists yet, and the plausible loss approaches whatever is present that day. That is why the sum insured informs poorly, and why two sites of the same contract value are not underwritten at the same price. The answers retaining a contractual value describe what the policy must pay at most, not what the underwriter fears.
Glossary entry →3. A site's exposure follows a curve. Where does it peak, and why is that moment the worst?
A few weeks before commissioning, at the very time of the testing period: the two peaks multiply rather than add
Maximum value and the first flow of energy through a plant never tested at load fall on the same day, and it is their coincidence that creates the danger rather than either alone: maximum value without testing would be a passive exposure, testing on a half-erected plant would cost less. Picturing the curve lets you place prevention measures where they act. The answer aiming at acceptance places the peak one notch too late, once testing has already happened.
Glossary entry →4. A natural event strikes a zone. What does it produce on a single site that would not happen across dispersed sites?
It mobilizes the main limit, several sub-limits and two deductibles at once
A natural event does not choose its targets within a zone: on a single site it reaches on the same day the works in progress, the site plant, the existing structures and the storage, that is, heads the policy bounds separately. You then discover that sub-limits each calibrated for an isolated loss all mobilize together. The other answers describe mechanisms independent of concentration, and two of them are wrong in themselves: the regime requires an official order, and delay cover requires covered physical damage.
Glossary entry →5. What is the most effective prevention measure on a large site, and what happens if it is changed without notice?
The supply schedule, and changing it without notice moves the subject from damage to disclosure of risk
Spreading deliveries flattens the exposure curve, which no protection measure does: the other answers name real and useful measures that mainly reduce frequency, whereas underwriting a site is dominated by severity. The second half matters as much as the first: bringing deliveries forward to catch up is a site decision that increases accumulation, therefore an aggravation to be disclosed, and nobody in the meeting where it is taken sees it that way.
Glossary entry →