When discussing the role of IoT in commercial construction it is not enough to imagine placing sensors throughout the jobsite. A modern commercial construction might involve thousands of pieces of data, derived from materials, equipment access, environment, information about devices themselves, and more.
There are certain engineering challenges inherent to getting that information to where it needs to be in a timely and accessible fashion.
A logical architecture could take the following form:
Devices → Connectivity → Edge → Data Platform → Analytics → Applications
Each element implies particular engineering decisions and there are multiple possible solutions at each stage. In what follows we will review potential considerations for each element in designing such an architecture.
- Determine Sources of Data
Before choosing the data acquisition devices it is important to understand the nature of information which needs to be obtained. A commercial construction IoT architecture might involve:
- RFID tags
- BLE & UWB devices (positioning, proximity sensors)
- GPS Devices
- Environmental sensors
- Machine telemetry
- Vision applications (cameras)
Each of these presents a different type of data and a different set of constraints which need to be taken into account including battery capacity, bandwidth capabilities, precision and more.
Therefore, it makes sense to determine the purpose of the IoT system in the construction field before moving on to other elements of the architecture.
- Connectivity Layer
With the list of required data sources established the engineering team can begin to think about how these devices will connect to the data platform and where each segment of the architecture will be hosted.
Depending on the construction site, connectivity can be established via Wi-Fi, cellular networks, gateways, BLE, UWB, LoRaWAN, and others.
This is also where considerations for bandwidth, latency, accuracy, power needs, and the number of required devices come into play.
It is entirely possible that for a large construction site multiple methods of connectivity will be involved.
- The Role of Edge
The data acquisition and collection layer may involve multiple steps before reaching the data platform.
This is where an application or infrastructure running on the edge can provide substantial value by performing filtering, aggregation, identification of interesting events, protocol conversion, temporary storage, and more.
An example of utilizing an edge gateway would be for instance in identifying important equipment events and only transmitting those to the data platform instead of all sensor information.
The use of the edge is merited in particular when dealing with bandwidth limitations or situations where latency is a concern.
- Designing the Data Platform
It is important to keep in mind that the overall purpose of an IoT architecture in construction is to collect and analyze data in order to gain actionable insights. Therefore, once all data has been transmitted and organized at the data platform it is necessary to think about how this information will be structured.
A data platform in its simplest form includes:
Device ingestion → Message broker → Processing → Database → Analytics/APIs
The specifics will depend on the unique requirements of the project. Regardless, it is important to make decisions about the representation of each element (devices, timestamps, location, etc.) to facilitate easier integration of data from various sources.
A consideration that often gets overlooked but can be quite impactful is ensuring the quality of data. Timestamps that do not make sense, duplicate events, omitted information for any given data point, or devices using the same identifier can render otherwise valuable information nonsensical.
- IoT in Construction: Integration with Other Systems
Typically, construction projects and management involve BIM systems as well as general business tools including ERP, CMMS, project control applications, building management, and more.
IoT integration with these tools is likely necessary to derive meaningful information and actionable events as many stakeholders will expect to find IoT data within their business application environment.
In other words, it is critical to think about IoT in the context of the broader ecosystem as opposed to designing a separate standalone solution.
- Security as a Priority
One of the advantages of applying a streamlined architecture is the ability to manage security functions at each segment.
In an IoT system, the number of devices which need to be secure grows exponentially.
Some of the considerations include:
- Authentication & Authorization
- Secrets and credentials management
- Encryption functions
- Segmentation and perimeter security
- Role-based access control
- Secure updates and patching
- Logging and auditing
- Protect data in-transit, at-rest, and in-use
Devices and their lifecycle should be part of this equation as well.
- Think About Sourcing and Scale
When first envisioning an IoT application in the construction field it is easy to assume that a handful of sensors will be enough to gather the information needed. In reality, a commercial construction project can easily involve thousands of devices.
This can present a serious architectural issue unless device provisioning, message volume, storage capacity, bandwidth, processing needs, APIs, and permissions are designed with large-scale operations in mind.
This is why it might be a good idea to begin with a single-purpose use case and then expand to additional applications and data sources as scale becomes an issue.
- Always Consider Failures and Offline Operation
In many ways an IoT application for a commercial construction project represents the inverse of an enterprise application. In the latter, the system is trusted to operate as designed and data integrity is assumed.
In a construction site however, there might be network outages, malfunctioning equipment, failing gateways, and more.
One of the primary responsibilities of an architect is to make decisions about the behavior of the application as it encounters failures. This could take the form of local storage, retry functions, monitoring, and logging for network interruptions, devices going offline, and unexpected behavior.
Depending on the use case certain offline or low connectivity procedures can be built in such a manner that the impact on operations or decision-making is minimized.
- From Data to Applications
When discussing applications it is necessary to keep in mind that an application in the context of IoT in construction can take many shapes and sizes depending on the needs of the project. There is likely to be considerable variation depending on whether the project involves heavy equipment or modular construction, for instance. Applications might focus on equipment utilization or access, materials, or specific aspects of the environment.
An important consideration involves not displaying data in and of itself but providing information and context on the construction site and project in general. What value does a mere stream of location data hold, for instance, if the application can tell whether a piece of equipment is being utilized?
A Practical Construction Site IoT Architecture
A commercial construction IoT architecture may include the following elements:
Physical Jobsite
↓
Sensors & Connected Equipment
↓
Connectivity / Gateways
↓
Edge Processing
↓
Message & Data Platform
↓
Storage + Analytics
↓
APIs / Applications
↓
Project Teams & Operational Systems
Each of the segments described above will have their own engineering, security, and application considerations. Designing the architecture in such a manner that these segments are distinguishable will contribute to an application which is easier to troubleshoot and expand.
Conclusion
Designing an IoT architecture involves a series of important decisions that cut across engineering, security, application design, data science, and more. Some of the considerations which should be reviewed when creating such an architecture for a commercial construction project include:
- Connectivity options
- Edge processing
- Data integrity and quality
- Security
- Integration with other systems
- Observability and failure scenarios
- Scale and reliability
- Information and application delivery
Creating an application which is focused and relevant to the jobsite involves understanding the constraints of each segment and working backward from the intended outcome. The primary value of an enterprise IoT application is not in the volume of data being collected but rather in information being extracted which can guide operations and increase productivity.
Top comments (0)