Data, and lots of it, drives today’s information-packed business climate. Effectively gathering and analyzing process and production data provides an ideal means to generate new revenue, improve operational efficiency, and streamline production in manufacturing and other industrial applications. Enabling this data accessibility is known as the Industrial Internet of Things (IIoT), which combines a number of contemporary automation, data exchange, and manufacturing technologies, marrying the concepts of Automation Technology (AT) and Information Technology (IT) to facilitate cloud-based access and data transfer.
Historically, cloud-connected technology has remained in the domain of the IT sector, freely connecting and transferring data through the Internet. As the needs of the AT sector have evolved, forward-thinking manufacturers continue to pursue avenues that optimize processes and make inroads into the adoption of “Big Data” concepts across the board. While many IT companies promote a definition of IIoT as the convergence of Operational Technology (OT) with IT, it’s possible to go one step further and say IIoT is also about IT converging with AT. Traditional automation technology, such as Programmable Logic Controllers (PLCs), require the adoption and integration of IT-oriented technologies. PC-based controls have been at the forefront of this change in the automation sector and continue to lead the pack with a rapid adoption of cloud connectivity protocols like MQTT.
The MQTT (Message Queue Telemetry Transport) protocol has become a standard in IoT communications. It is an extremely lightweight, publish-and-subscribe protocol currently used in many popular IoT applications and millions of devices. Data quality is important, so a Quality of Service (QOS) mechanism is integrated to ensure properly received data. As a result, MQTT is an ideal protocol for IoT and M2M applications.
The publisher/subscriber mechanism means that the device sending the data is “pushing” information to a message broker. The message broker is implemented in software and serves as a secure, central hub the messages pass through. This central hub can be located on a local PC, an enterprise server or in the cloud. Large cloud service providers such as Amazon Web Services also offer broker capabilities. The messages that are pushed to the broker are transferred to a certain user-defined topic. These topics are organized into levels. For example, a user can send a value of 32.3 to topic “Pump1/Pressure” and send another value to “Pump1/Temperature.”
Clients interested in the data can subscribe to relevant topics like a SCADA system, mobile HMI or separate controller. When the publisher sends data to the broker with a given topic, the broker receives the data and sends it to all subscribed devices. The subscribers can subscribe to the data, then drop the subscription without any changes to the publisher. The data is then quickly accessible from multiple places and doesn’t require time to configure the publisher to be aware of the data consumers.
QOS mechanisms ensure data has been successfully transferred to all intended subscribers. Within MQTT, there are three levels of QOS starting with level 0. As the level increases, so does the reliability of message delivery, adding additional (though minimal) latency and bandwidth requirements. At QOS=0, the message is only delivered to the subscriber once with the confirmation. At QOS=1, the broker will deliver the message at least once and will require a confirmation of receipt. Level 2 means that the broker will deliver the message exactly once but will do so using a four-step handshake method to ensure delivery.
Modern control systems enable users to easily send data using known PLC programing elements such as function blocks, allowing programmers to quickly implement IoT protocols. With this unfettered access to the data, information can be supplied to an analytics server located either in the local network or in a cloud network. Machine operation optimizations are additional benefits of analyzing the data and uncovering insights. To achieve quick gains from machine data, a basic analysis of machine operations is performed. This includes processes such as cycle time differences, i.e. how long a machine may stay in one state or another, or even overall output throughout the day.
Data analytics has helped manufacturers and device end users optimize their processes and procedures, but cloud technology and IoT communications open up new possibilities for original equipment manufacturers (OEMs). OEMs are cloud-enabling their devices to publish data if the end user has enabled web access on the equipment. This equipment then sends data such as common alarms, uptime versus downtime, and common settings. With this data, the OEMs gain better insight into how effectively equipment is being used and the most common issues that arise with their capital equipment. IoT communications and data analytics drive new product designs and facilitate clearer focus so companies can best implement truly impactful production solutions that maximize value and profitability.