Construction sites may be physical environments, but many of the challenges that teams face are deeply tied to data.
The questions range from where is this piece of equipment, which materials have been delivered, where are our workers operating, how is a project progressing across various work zones, etc.
IoT technology can help provide answers to these questions by connecting physical objects and environments to ones digital. AI can then take it a step further by analyzing this data.
This combination is commonly referred to as AIoT: AI + IoT.
AIoT Data Flow
An AIoT system can be thought of as a series of steps:
Physical Environment → Sensors → Connectivity → Data → AI → Insights → Action
Every step in the sequence plays an important role.
Sensors and identification technologies are used to capture information from the physical environment. This can depend on the use case, but can involve RFID, BLE, UWB, GPS and connected sensors among other technologies.
Connectivity is about getting this information somewhere where it can be processed and stored.
Processing can then combine this information, from different sources, together. Information from various sources can be related to form a greater context, rather than treating each entry individually. The data would include information about where something was located, how it moved, what equipment was involved, what materials were needed, and what activities were done on a jobsite.
AI can then analyze this data. This could be looking for insights, patterns, oddities, or anything else that would allow for an analysis of the data. Finally, this information would be turned into actionable insights that could influence a decision about something on the jobsite.
Why Construction is a Good Application of AIoT
Some of the biggest challenges that construction sites face involve constantly changing environments.
Equipment moves between work areas, materials are delivered throughout a jobsite, workers are operating across different zones, and project conditions change throughout the jobsite.
Being able to understand a situation in real-time, and in the context of how things have been in the past, can provide value.
For instance, RFID can provide identification information about tagged assets or materials. GPS can provide location information of equipment in outdoor environments. BLE can provide proximity information, while UWB can provide more accurate positioning information in particular environments.
Connected sensors can also provide more information about equipment or the environment that they're in.
The information from these different sources don't need to be used individually: they can be combined together for more data depth.
Combining these Data Sources
Let's take a look at an equipment tracking scenario.
A location system would provide information on where a particular piece of equipment is. A utilization system would provide what activity the piece of equipment is doing. A project management system would provide information of what worksite currently needs the equipment.
When they're put together, this allows for a more complete view of equipment utilization on a jobsite. The same principles apply to materials and workforce information.
Material records can be combined with where the materials are and the current activity of a jobsite. Workforce information can be considered with work zones and the activity on the jobsite.
This data can be analyzed by AI to help create connections rather than relying on people to recognize patterns.
Creating Useful Construction Intelligence
The goal isn't about creating as much data as possible. It's about approaching an operational challenge and determining what data would be needed to solve it.
If asset identification is important for a use case, then RFID is appropriate. If knowing about the location of an asset outdoors is important, then GPS may make sense. If proximity information is important, BLE may be the right choice. If more accurate positioning is important, UWB may need to be considered. If information about equipment or the environment is important, connected sensors can provide the necessary information.
It's about determining what technology (or technologies) to use, based on the asset, environment, accuracy, and approach that makes sense for a given use case.
For a more construction specific example of applying AIoT technologies to workforce, equipment, materials, access, and project activity, I've provided more details at CommCon AI.
The Developer View
The reason why I personally find AIoT so interesting is that it's an intersection of several systems:
Physical world → Data collection → Integration → Analytics → AI → Operational systems
Beyond building an AI model, data quality, device connectivity, identification, positioning, integration, and system context are all important variables that affect what value can be derived from the data.
It's especially interesting in the context of construction since the physical world is constantly changing.
Ultimately, AIoT can allow for a connection between what happens on a jobsite and the digital world where analysis takes place. The objective is simple: be able to convert connected physical world data into information that can be used.
Top comments (0)