How AI will influence the evolution of the traditional surveying toolkit:
In 2017, the value of the world’s real estate reached a staggering $280 trillion. Central to the valuation process behind this figure was the work done by thousands of surveyors around the world. Although their profession often flies under the radar, their task – analysing properties and producing official valuation reports – is necessary for various actors within the real estate sector, from banks and lawyers to construction workers and city planners.
Becoming a surveyor isn’t easy. It takes roughly 3 to 4 years to complete the training and achieve chartered status. Prospective surveyors will need to master certain equipment, and learn about the variables affecting a given property, including floor plans, soundings, and building conditions.
The many moving parts within surveying projects pose two main problems. Firstly, the sheer quantity of variables and data points to consider makes the process incredibly time consuming. Even before physically assessing the property, surveyors spend a day conducting desktop research, producing a detailed preliminary report consisting of Google Street view searches, analysis of public data comparables, data on local points of interest, and historical data from property portals. Coupled with the high level of practical skills required to qualify as a surveyor, the process is indeed an expensive one. Secondly, as reports are only valid for a limited period, the cost to the consumer is not only high, but also frequent
In this era of constant technological innovation, industries from web design to agriculture have leveraged the power of AI to reduce their reliance on inefficient manual processes. Yet despite this, the tripods and theodolites of traditional surveyors still feel all too common. However, the idea that data collection processes (such as those used to produce preliminary reports) could be integrated into a single time-saving algorithm is hardly revolutionary. In this context, reports could be finalised within minutes, and confirmed by the surveyor simply by double checking the information and validating the evidence.
Certain platforms have already implemented these concepts. Vigilant adherence to certain rules and regulations is central to surveying, and data-driven technology has indeed been used to facilitate this. AI Solutions, for example, has streamlined asbestos monitoring, their software enabling users to collaborate on crucial data points, apply the data to similar properties, automatically screen their contractors’ competence, and simplify the communication of the resulting reports.
The same AI-powered software that’s managing asbestos could perhaps use smart sensors to capture other real time data, already proved possible with leak sensors. Even difficult, dangerous, and often impossible roof inspections, using data obtained from a drone’s imaging software shared instantly with every other surveyor in the area, could become risk-free.
If software like BIM can help create a building using a single system of computer models, a future in which smart sensors and machine learning technology, integrated into the building process through apps that take real-time data from a nationwide network of surveyors, doesn’t seem far-fetched. Reduced errors, rework and cost, for both consumers and businesses, seem technologically feasible. You might imagine platforms like Seeable working alongside advanced 3D imaging technology like this piece from Leica.
The surveyor of the future will still have similar requirements. They’ll still need to know how to assess a building’s condition and learn the laws regulating their profession. But in their pocket, you might find the latest smartphone, containing an app that connects them to the combined knowledge of both the latest machine learning algorithms and the professional experience of their human peers, enabling them to complete their job more quickly and accurately than their predecessors. While they’d still manually capture images, video, and sound, the AI system analysing it in the background could generate a custom checklist of the relevant tasks to be executed, auto-populate all relevant data from public sources, use OCR analysis to highlight relevant parts of the property, and provide a collaborative note-taking platform, so that by the time the surveyor returns to their office, all that’s needed is to put the pieces together.
Yet, it seems every company wants to build their own thing. It’s said that AI could theoretically do 90% of a surveyor’s work, but the beauty of the human surveyor is their ability to connect the dots throughout different stages of the problem. AI Solutions wants to facilitate asbestos management, not determine the leakiness of a roof. Seeable creates 3D representations of interiors; Google’s Street View shows the outside and it allows the surveyor to go back in time via the archived street view images. We have the technology to quantify and process almost every moving part and minimise human risk. Companies like Google’s AI platform might be a start, emphasizing collaboration and flexibility, but while machine learning is indeed capable of comprehending systems of laws and processes, what it will advise – and how it will get to those decisions, is a process we still have yet to decode. What we do know, though, is that we need innovative software solutions that are affordable and able to integrate with existing smartphone hardware technologies.
