Sensors and IIoT device implementation
Sensors and networks are important to Industrial Internet of Things (IIoT) implementations, according to experts at the Sensors Midwest conference. See how to turn sensor data into actions in four steps.
How sensors and networks are enabling Internet of Things (IoT) implementations across industries were among topics and case studies discussed at the Sensors Midwest Conference at the Donald E. Stephens Convention Center in Rosemont, Ill.
"You really need IoT skills to work with the right sensors, database skills, and process fulfillment to bring it together and increase the value to our partners," said Scott Schwalbe, CEO of NimbeLink and keynote speaker.
Schwalbe also stressed that the role sensors play in the IoT space is significant and that organizations need to think about how they’re going to tackle an IoT implementation and a solution for customers. Currently, only 26% of companies have a successful IoT launch. Do they have the internal resources to do it or does it need to be outsourced? Schwalbe said to fail big, fail small, and fail fast.
The conference session, with an IIoT University theme, maintained the momentum of the IoT and the sensors discussions, and included cybersecurity issues, how wireless sensors are collecting data and creating actionable data with IoT networks, and use cases for embedded sensors.
Experts weighed in on factors that need to be considered with IIoT devices. One of the biggest issues creating an IoT network is ensuring that the network is secure. One of the biggest fears for many companies, if not the biggest, is leaving proprietary data open to a cyberattack.
Vinai Sundaram, co-founder and CEO of SensorHound Inc., spoke about monitoring IoT devices after an IoT implementation. Many cyberattacks have been publicized; it can be devastating to a company and its customers to expose critical data. It was noted that currently, there are 8.4 billion IoT devices, and by 2020 there will be a projected 22 billion devices. Sundaram said securing IoT devices is not easy, and that there are several layers of complexities to work through to implement effective cybersecurity, which is a big reason why many companies have not implemented IoT architectures yet.
Many companies still are unaware of what steps need to be taken to implement IoT offerings, mentioning that they don’t have enough internal resources, and don’t know what their data is capable of and what is needed in a solution. Cybersecurity is a complex issue that many companies have yet to grasp to prevent cyberattacks. If the complexity of IoT devices isn’t thought through, a company can leave itself vulnerable to a data leak, among other types of attacks at the network, firmware, and hardware levels. It critical to have security measures in place, as well as proper monitoring.
"It’s like locking the door but leaving the keys underneath the doormat. How you manage the secret key is critical. You need a constant monitoring solution to identify if IoT devices are functioning properly," said Sundaram. IIoT device security is difficult for many reasons including:
- Resource constraints (lightweight cryptography and intrusion detection)
- Secret key management (identity, authentication, and authorization)
- Security configuration and policy management.
While cybersecurity is difficult to implement fully, ways to help prevent cyberattacks include network intrusion detection, secure networking, edge intrusion detection, and lightweight cryptography.
Turn data into actions: Four steps
Another issue is what to do with sensor data. Murali Kashaboina, chief data scientist at Entrigna, broke down the steps on how to turn collected data into actionable data during one of the technical sessions.
Step 1: Locate the data you’re collecting. If you don’t know the characteristics of the data, the outcome may be less than optimal.
Step 2: Understand what the data means. Review the process that touches the collected data. Talk to all of the people who use the systems to better understand data sources.
Step 3: Identify the next step. What do you want to do next? Brainstorm action items that would benefit the business (such as improve workflows or increase inventory returns). The focus needs to be on prescriptive, rather than predictive, actions. What are the "missed" actions?
Step 4: Implement in small steps. Don’t get into analysis paralysis, said Kashaboina. Start small. There’s lower risk and it costs less if you can take corrective action right away and build upon that solution for a larger project down the line.
Although the industry is still experiencing some growing pains with IoT devices at varying stages including implementation, creating viable cybersecurity measures, and figuring out what to do with collected data, it’s apparent how much potential IIoT devices have across many industries. Experts presented applications that will have a significant impact for many industries and the environment.
For example, the City of Chicago can use collected data from sensors to optimize services such as water management, waste removal, health inspections, and to reduce traffic incidents. Projects also are testing the world of "digital agriculture" and creating wireless underground sensor networks for irrigation. It’s not a matter of if IIoT devices will be used, it’s a matter of when and how various industries will use these devices to resolve issues and maximize return-on-investment.
Emily Guenther is an associate content manager for CFE Media, firstname.lastname@example.org.
- Factors to consider for a successful IIoT solution
- The importance of cybersecurity on three levels within the network
- Emergence and growth of IIoT in the sensors marketplace.
How should organizations assess whether or not they should outsource their IIoT implementation and maintenance?
Control Engineering has online pages for IIoT and sensors.