Commercial construction sites act like decentralized physical systems.
Personnel move around the site, equipment is constantly relocated, materials are brought in, staged, and eventually installed. Access points log credential events. On-site inspections and task completion points yield various additional pieces of data.
The real challenge is that this data and events are isolated within specific systems and cannot provide contextualized insights in real time.
AIoT can be defined as an architecture for integrating these physical events with software intelligence to enable real-time actions.
The Basic AIoT Architecture
A helpful way to think about AIoT in the construction environment is like the following architecture diagram:
``text
Workers / Equipment / Materials
RFID / BLE / UWB / GPS / Sensors
Edge & Network Layer
Event Normalization
AI / ML Analytics
Operational Applications
`
The most important element here is not the individual technology itself, rather how events and data interact.
The combination of a simple GPS read versus equipment specifications reveals two distinct parts of a system: your equipment. Your project record displays where the equipment is assumed to be, your schedule shows when the equipment should be somewhere and working, thus combining these datasets provides valuable operational context.
Why We Need Multiple Location Technologies
Construction environments differ from one site to another. Mobile assets can benefit greatly from GPS in an outside area while tagged assets like materials or tools might operate more efficiently using RFID or BLE technology. Higher precision localization can even benefit from utilizing UWB technology for more accurate proximity results.
Given this, your ideal application may use multiple technologies at once versus trying to incorporate just one single technology throughout your platform. For example:
`text
Outdoor fleet - GPS
Tools / assets - BLE / RFID
Known, fixed location points - RFID
Highly accurate tracking - UWB
Long range low power - LoRaWAN
`
Selecting the appropriate technology all depends on how you plan to use your system considering its individual attributes such as accuracy required, coverage area, ease of mobility, and required infrastructure.
Critical Use of Event Normalization
These data points below reflect a series of actual events on-site:
`text
Badge A Gate 2 07:41
BLE Tag Floor 12 08:03
RFID Tag Material Yard 08:17
GPS Asset Project B 08:30
`
Simply looking at them will not result in usable, operational information.
Instead, a dedicated data layer for processing will help normalize those event into usable records for your AI application. This typically involves a simple mapping to fields like:
`text
identity
asset_id
location
timestamp
event_type
project
zone
authorization
status
`
Once standardized like this, you can integrate artificial intelligence to start taking meaning from the correlation between events and data points such as worker access events with permission rules and expected work schedules, comparing the actual location of an asset against the desired project status and estimated time of usage, or analyzing materials moved across a project versus the purchasing records.
Workforce Analytics
The possibility to track employee presence through a connected, smart site is an especially compelling advantage for commercial construction where teams on-site may vary widely by contractor and specialized trade with workers moving between different areas on-site, often up multiple floors. Information can be gathered on:
How many workers are at the site currently?
How many workers are present in a specified location area on-site?
How long workers remain in each respective zone on-site?
How many teams are at the site concurrently?
Worker movement throughout their assigned work shift.
Worker activity at building access points.
An AI analysis is capable of contextualizing such data against work schedules and an estimation of predicted task completion timelines, with the core goal of helping project teams visualize labor on-site and identify work front issues rather than to micromanage staff.
Equipment Analytics
An industrial tracking architecture can also be utilized with machinery. Imagine you have hundreds of various assets across many construction sites. You can abandon only tracking them based on manual entries in a ledger. Instead, by collecting utilization information and location data, you can begin to create a system for analyzing machine history:
`text
Asset Location Project Utilization Movement
`
Through these historical records, you will be able to answer many pressing questions such as: is my equipment being used, where was it last seen, is it at the correct location according to project specification, is it unnecessarily idle on-site, and should it be relocated to a different site? This is one of many areas where AI analytics can add much value beyond simple tracking.
Material Intelligence
Materials also provide an additional level of important data – movement throughout a workflow.
`text
Supplier
Receiving
Laydown Yard
Temporary Storage
Work Zone
Installation
`
By collecting movement data as materials go through each critical stage of the supply chain and construction workflow with an automated identity system, project teams can create more accurate and reliable records regarding material tracking. This data, analyzed with AI tools, can help connect all aspects of a construction project together; Procurement, Logistics, Warehousing and Field Operations can better monitor materials flow.
Integrating AIoT with Existing Construction Software
The most valuable AIoT applications and information must not stay in their own separate application isolated from your existing systems. The true benefit of a smart connected solution comes from integrating with existing building information modeling (BIM), enterprise resource planning (ERP), Computerized Maintenance Management Systems (CMMS), scheduling, procurement, workforce management and more.
A very basic example of what an optimized and connected AIoT construction architecture looks like may be similar to the diagram below:
`text
IoT Devices
Edge / Connectivity
Event Processing
AI Layer
BIM ERP CMMS
Project Controls
Operational Action
``
CommCon AI states AIoT for commercial construction focuses on workers, site access, equipment, tools, materials, and the speed at which construction is completed and offers extensive information on AI Functions for AIoT-enabled Commercial Construction here: https://commconai.com/ai-function-for-aiot-enabled-commercial-construction/ where all this data can come together.
Starting with the Workflow, Not the Hardware
One pitfall when starting any connected construction project is immediately asking the question: "What type of IoT sensor should I buy and then how can I use it?" A more beneficial question would be instead: "What aspect of the current operation can I make more efficient with this data." Technology selection comes easily once you identify the real need on site.
An outdoor site may need GPS while a workshop may need something closer range like BLE and an area you want extremely accurate tracking within may require UWB sensors.
Technology decisions should follow what can be optimized.
Last thought
Your jobsite is already creating value and producing a large amount of data. Your primary role today then becomes managing the integration of various physical events and producing truly actionable information. AIoT offers a strategy to begin integrating this, it allows you to capture, interpret, normalize and begin to relate these specific physical events against all other project requirements and provide relevant results to the decision makers. Connecting the physical world to an intelligent realm of construction.
For More Vist: commconai.com
Top comments (0)