Construction data
The demand for hosting of construction data is only going to increase as the technology supporting the industry innovates, examples being:
- Electronic 3 and 4 D designs
- BIM
- Drone surveys
- Planned and actual programmes
- Project management / contract management software
- Contract and actual cost data.
- Artificial intelligence.
Consequently, should a claim situation arise there is the increasing potential for the claimant to submit voluminous supporting evidence which could quickly become disproportionate and costly for all parties.
Could sampling be a solution?
What is statistical sampling?
This methodology can provide a reliable and efficient way to assess construction claims and make informed decisions based on making an unmanageable volume of data manageable.
Attempting to keep this overview simple, this sampling would involve:
- Scope confirmation
- Extent of relevant data
- Agreement of sampling method (random / stratified / systematic / cluster)
- Based upon the desired confidence level and margin of error, the use of statistical formulae to calculate an appropriate sample size.
A practical example would be the review of 10,000 data sources, with a required confidence level of 95%, an error rate of 5% and a margin of error of 2%.
Statistically the number of data sources required to be reviewed would be 450, which as 4.5% of the total data set of 10,000 is a significant reduction.
Where has it been used?
Examples of the use of a sampling methodology include:
- Voluminous contract variations
- Multiple defects
- Thousands of activities in relation to delays
- Quality control including cladding inspections
What is the view of the courts?
Building Design Partnership Ltd v Standard Life Assurance Ltd [2021] EWCA Civ 1793.
The UK Court of Appeal in this case upheld the use of extrapolation to plead claims, and noted that the days of the court requiring parties in detailed commercial and construction cases to plead out everything to the nth degree were over. Such an approach was not sensible, cost-effective or proportionate. Instead, parties were duty-bound to find a way of trying out the principal issues between them in a sensible and proportionate manner.
What are the potential issues and solutions?
- Avoid sampling errors, which could represent fundamental issues for claim success.
- Avoid sampling bias, resist the temptation to be beneficial for the client.
- Use a statistics expert.
- Agree with all parties the proposed sampling methodology.
Summary
There is no doubt that the amount of construction data available to support construction claims is going to significantly increase. We are at the start of a statistical revolution.
Courts see benefit in introducing proportionality.
The key to success will be to implementing an appropriate sampling methodology.
Contact our experts if you require any advice or guidance.
