Weather predictions are critically important for many sectors of the economy. The science behind these predictions relies on the use of computers to extrapolate data using complicated atmospheric physics. Nevertheless, multi-day outlooks cannot be wholly relied upon. In fact, NAV Canada’s forecasts for aviation, an industry dependant on accuracy, do not extend beyond 12 hours.
This might soon change with the help of artificial intelligence (AI), and construction could be a big beneficiary.
UK-based software developer MetSwift believes its weather prediction programmes based on AI can integrate new levels of weather analysis into business processes spanning several industries. The implications for the construction industry are important,” the company says. “Extreme weather is a significant contributor to loses in construction. Costs associated with weather are as high as 21 per cent.”
According to MetSwift CEO James Banasik, weather is beginning to overlap with insurance. Accurate predictions and quantifying of weather risks is gaining prominence among insurance professionals and will require a response from construction.
Banasik says that the construction industry, although bigger than the aerospace and automotive industries, “has been trapped in a cycle of low innovation and productivity”. Given the impact weather can have on timelines and therefore project costs, addressing weather in an integrated, predictive manner is becoming important.
Attempts to accurately predict the weather go a long way back. Admiral Robert FitzRoy, the man regarded as the father of weather forecasting — a term he himself created — began issuing two-day weather outlooks and storm warnings in the early 1860s as head of the British Meteorological Department. These were published in The Times and became popular for predicting the weather for sporting events such as The Derby. However, FitzRoy faced scepticism, even mockery, from the scientific community when predictions failed to be realized.
It’s not that weather forecasts aren’t being considered in project management already. However, MetSwift says that the current weakness lies in the integration of forecasts into BIM platforms beyond the estimates inputted by project managers. “There is no forecasting element actively integrated into 4D or 5D BIM, meaning this data is of limited value in design and planning phases.”
Using AI algorithms developed from data gathered from over 92,000 weather stations around the globe, MetSwift believes it has the world’s fastest peril risk calculator, giving it the capability to forecast well beyond the limited horizons of current weather models. Their AI programming identifies specific weather events and then tracks their lifecycles and interactions. This huge body of information is weighted under scenarios that can, so the company claims, calculate their effect in terms of probability up to five years in advance. “These predictions can be easily incorporated into BIM instantly.”
“Once uploaded, various exact ‘what if’ scenarios are posed,” the company says. “From these, real-time, time cost projections are determined.” MetSwift offers a simple example: Is it worth the extra cost of increasing the workforce and overtime in order to enclose the structure sooner? The answer depends on the probability of extreme weather inhibiting work being done while exposed to the elements.
AI is changing the way construction industry partners design and build in many ways, according to Professor Peter Debney, Application Specialist with global engineering software developer Oasys Software. While the increased use of AI in robotics used to outperform humans in repetitive tasks often comes first to mind, Debney also points out that, “AI systems can collate and organize information for engineers to use within project planning and design implementation.”
Planning, project management, communication during the build and post-construction are all witnessing increased investment in AI developments that are being integrated into BIM platforms, says MetSwift. The company believes, however, that its predictive AI software is a “game-changing paradigm shift in 4D and 5D BIM. Real-time changes in time and cost analysis that incorporates weather prediction are around the corner.”
John Bleasby is a Coldwater, Ont. based freelance writer. Send comments and Inside Innovation column ideas to firstname.lastname@example.org.