Risk analysis makes it possible to estimate the probability and consequences of incidents that may affect a project. It is an essential part of planning: it does not prevent unforeseen events from happening, but it helps to minimise them, react quickly when they appear and make more informed decisions. In practice, there are two complementary approaches.

Qualitative analysis

This is the starting point in 99 percent of projects: fast, simple and sufficient in most cases. It consists of classifying each risk in a probability-impact matrix, for example using scales from 1 to 5, where the risk score is the product of both. The result feeds the risk register, which prioritises where attention should be focused. Its weakness is subjectivity: it depends on the judgement and biases of the person assessing it.

Quantitative analysis

When the risk is high or the decision is critical, such as a go/no-go decision, a contingency reserve or a date commitment, probability and impact should be translated into numbers. In construction, the most useful techniques are:

  • Three-point estimation, using optimistic, most likely and pessimistic values for durations and costs.
  • Monte Carlo simulation on the schedule and budget, to obtain a finish date and cost with an associated probability, the well-known P50 or P80, instead of a single number.
  • Expected monetary value, or EMV, to size contingency reserves.
  • Sensitivity analysis, using a tornado diagram, to identify which risks really move the needle.

The quantitative approach is more objective and easier to present to management, but it has an Achilles heel: it needs quality data. The most common problem is not the technique, but the lack of sufficient reliable data to analyse. Without a good schedule model and historical production data, the resulting number is as subjective as a qualitative estimate, only with the appearance of precision.

The practical rule is to always start with the qualitative approach and reserve the quantitative approach for the risks that truly justify it.

The most common causes of delay

A good risk register starts by understanding where projects usually fail. In construction, delays are repeated because of:

  • Incomplete or late design and engineering, and continuous scope changes.
  • Permits, licences and expropriations that take longer than expected.
  • Supply of materials and equipment with long lead times.
  • Low productivity or lack of qualified labour.
  • An unrealistic or poorly sequenced schedule from the outset.
  • Interferences between disciplines due to lack of coordination.
  • Unforeseen ground conditions, such as geotechnical issues, and weather.
  • Slow decisions and approvals by the client.
  • Rework caused by quality defects.

The most common causes of cost overrun

Many cost overruns are, in reality, delays under another name. The most frequent are:

  • Poor initial estimating and budgets with optimistic bias.
  • Scope changes and uncontrolled variations.
  • The delays themselves, which generate indirect costs and extensions.
  • Volatility in material and energy prices.
  • Rework and quality defects.
  • Unforeseen ground conditions.
  • Poor contractual and claims management.
  • Contingency calculated incorrectly because risk has not been quantified.

How we approach it at PlanBuilt

At PlanBuilt, we integrate risk analysis where it has real consequences: in the schedule. We run schedule risk analysis with Monte Carlo simulation on the P6 model to provide dates and contingencies with associated probability, and we connect the risk register to real progress in an interactive online dashboard, so risk monitoring stops being an Excel file nobody looks at and becomes part of the project's daily control.