Automation systems generate large amounts of data that can be used to improve business performance, but are often unavailable and not efficiently managed and analyzed. Unfortunately, this valuable data is often inaccessible for a number of technical and economic reasons. Newer architectures are changing this predicament by combining the flexible and powerful edge computing model with a cloud computing model, making it not only feasible but practical to analyze data, gain new insights and make the results available to stakeholders. This article highlights some of the most common reasons why organizations struggle with untapped data and how a modern edge-to-cloud IIoT solution makes it easier to access and use that data.

How the infrastructure generates unused data

Until recently, most production data was sourced from PLC, HMI, SCADA and Historian systems running in OT (Operations Technology). These systems are focused on control and visibility to maximize operational efficiency and available uptime. The use and analysis of associated data that goes beyond immediate production goals is secondary.

The OT infrastructure was designed and scaled according to operational requirements, which resulted in the following decisions, for example: Selecting proprietary communication protocols that meet performance requirements, but without flexibility and cross-vendor compatibility. Minimizing the collection of control and sensor data to increase system reliability and simplicity. Implementing local architectures on site to minimize cybersecurity threats and vendor lockout programs to protect intellectual property and increase reliable machine operation, but often at the expense of connectivity.

The corresponding systems function perfectly in the context of their operational objectives, but have data weaknesses and do not benefit from analyzing all potentially available data.

Within the OT environment, data sources appear to be open, but in reality, applications outside the OT environment, where the data could be more easily analyzed, have difficulty accessing it. In addition, many potentially valuable data sources - such as environmental conditions, condition monitoring information and energy consumption - are not needed for production or plant control and are therefore not captured by automation systems. The possibilities for analyzing large amounts of data are increasing, but the limited availability of unused data remains an obstacle.

Types of unused data

There are many forms of unused data that come from machines, the factory floor and other systems that are part of the OT or provide balance to the plant. This data can be as detailed as a single temperature reading or as extensive as a historical data log showing the number of acknowledged alarms by an operator. Typical unused data is:

Isolated assets in a plant without network access to an OT or IT system. This is the simplest case, but not necessarily the easiest to solve. Take a standalone temperature transmitter with 4-20 mA connectivity or even Modbus functionality. It needs to be connected to some kind of edge device - PLC, edge controller, gateway or similar - to make this data stream accessible. In many cases, the data is not important for machine control, so it is not available in traditional, legacy PLC/SCADA data sources. Pulling the data through the nearest machine PLC risks voiding OEM warranties due to required changes in program logic.

Unrecognized assets that are connected to OT systems and generate data but are not used. Many smart edge devices provide basic and advanced data. A smart voltage monitoring device can provide basic information such as volts, amps, kilowatts, kilowatt-hours, etc. using wired or industrial communication protocols. But more detailed data sets such as Total Harmonic Distortion (THD) cannot be transmitted due to lack of application requirements, low bandwidth communication or limited storage capacity of system data. Although the data is available, it is not used.

Assets that generate data but are queried at an insufficient data rate. Even if a smart device provides data to monitoring systems via a communication bus, the polling rate may be too low or the latency or data set too large, so that no useful results are obtained. Sometimes the data is summarized before it is published, resulting in a loss of accuracy.

Inaccessible assets that generate data (often not process data, but still data that is important for diagnostic purposes), but in a format that is not generally accessible or available through traditional industrial systems. Some smart devices have built-in data such as error logs that cannot be transmitted via standard communication protocols, but would still be very useful in analyzing events that have led to downtime.

Non-digitized data - Personnel create data manually on paper, clipboards and whiteboards, missing the opportunity to digitize this information. At many organizations, employees fill out paper test and inspection forms and other similar quality documentation without making provisions to incorporate this information into digital records. A more modern approach uses digital methods to capture this data, leading to a 'paperless plant'.

Added value through data in the cloud

Unused data is of significant importance for companies that want to analyze the operational performance of the entire production plant or multiple plants. They are looking for solutions to transfer unused data from the field to the cloud for logging, visualization, processing and more detailed analysis. This connection, especially to powerful IT systems - onsite or cloud-based - is needed so that many types of edge data can be stored and analyzed to achieve more detailed and longer-term analytical results. This goes far beyond what is normally done for short-term production-oriented goals.

If an end user or OEM can free unused data from traditional data sources and transfer it to cloud applications or services, many possibilities are created such as remote monitoring, predictive diagnostics and root cause analysis, cross-machine, cross-plant and cross-plant planning, long-term data analysis, comparable asset analysis within the plant and across multiple plants, machine park management, cross-functional data analysis and analytics (deep learning), insights into production bottlenecks and identification of the origin of process errors, even if they are only detected later in the production process.

Edge connectivity solutions

The purpose of IIoT initiatives is to solve problems related to unused data and effectively connect edge data to the cloud where it can be analyzed. IIoT solutions include hardware technologies in the field, software running both at the plant edge and in the cloud, and communication protocols, all effectively integrated and built to securely and efficiently transmit data for analysis or other purposes. Edge connectivity solutions can be a component of automation systems or installed in parallel to monitor data that the automation systems do not need. They

can take many forms, including compact or large PLCs connected to industrial PCs (IPCs), edge controllers running SCADA or edge software suites, and IPCs running SCADA or edge software suites. Once the data is received, it must be pre-processed or at least organized by adding context. Context is particularly important in a manufacturing environment, where hundreds or thousands of digital sensors monitor mechanical and physical machine operations.

Finally, the data must be transferred to higher-level systems that use protocols such as MQTT or OPC UA. Today, OT/IT standards are developing in a way that ensures the consistency and future flexibility of data and communication.

With any solution, it is important that it is flexible but also meets the standard - unlike individual setups, which are impossible to maintain in the long term. Once an edge solution is set up and receiving data, the next step is to make it available to higher-level IT systems - via seamless communication with the software in the cloud. Hosting software in the cloud offers a number of benefits including lower costs as the user only pays for what they use and avoids investment in the purchase and management of IT infrastructure. Cloud computing is often referred to as an "elastic computing environment", as more computing power and additional data sources can be made available in real time if required. The cloud also eliminates problems for the user such as the configuration of IT hardware and software systems as well as their deployment, management, performance, security and updates.

Essentially, the cloud provides the ability to efficiently process large data sets with CPU performance scaling based on analytics requirements. Data security can be enhanced with different servers for storage with back-up and disaster recovery options. A cloud architecture is particularly suited to the requirements of companies implementing IIoT data projects. The cloud provides the necessary infrastructure for many IIoT projects, and the combination of these two technologies enables innovative interactions between people, objects and machines and is the birth of new business models based on intelligent products and services.

Silvia Gonzalez, Director of Product Management, Emerson

SPS: Hall 7, Stand 490