The Internet of Things (IoT) & Robotic Process Automation (RPA)

Andy Slote - Director of Customer Success for ObjectSpectrum

September 9, 2024

Robotic Process Automation (RPA) is a commonly used technology for business process automation. What may not be immediately obvious is that businesses implementing RPA solutions should also consider incorporating Internet of Things (IoT) technologies. Do these two strategies work effectively in the same environments? Are there elements of each that conflict, or are they complementary?


What is RPA?

RPA is mainly an automation technology that allows for creating “bots” (that’s the “robotic” part) to perform repetitive, mundane tasks, particularly involving the exchange of data between various data-oriented systems. A human user typically carries out these tasks on a computer.

One typical example is automating data entry by capturing and replicating a human user’s actions on an online interface or GUI. This capability can also be used to capture data from a screen or window, known as “screen scraping,” and convert it into a format that can be input into another application. Rules can be set up to verify the information before it is entered and to perform calculations or transformations on the data. Advanced bots can be developed to navigate multiple screens, clicking icons or buttons as needed, to imitate the manual human process.

RPA platforms enable the use of Application Program Interfaces (APIs) when available, providing a software interface to access data from a system. Conversely, RPA can create an external API when a legacy system does not support it. An RPA implementation could use screen scraping to capture information from a legacy system that lacks an API while simultaneously obtaining information from another source using an API, merging the data from the two sources, and providing the result to a third system via its API.

You can create complex logic that interfaces with different systems—some serving as inputs, outputs, or intermediate steps. This might involve intentionally timed delays, rate limiting, queuing, and distributing transactions across multiple systems to balance the load or for functional and geographic routing.

RPA platforms often incorporate Machine Learning (ML) and Artificial Intelligence (AI) by storing information in data repositories for ML/AI access or making it available to ML/AI applications through APIs. ML/AI interprets data, recognizes patterns, and further automates the processes.

Another common RPA feature is the automatic generation of communications via email, notifications, and chatbots. These communications can be triggered based on simple rules.

Anything that can be defined by a set of business rules involving tasks that can be accomplished by interfacing with various systems is fair game. Some of those rules may involve decisions made by human users. In the past, these decisions were typically simple ones based on rules and lookup tables. However, as machine learning/artificial intelligence becomes more sophisticated and capable, more and more of those decisions can be automated.

One major driver of RPA is the cost savings achieved through automated processes, which require fewer human users to perform mundane and repetitive manual tasks. In addition to cost savings, RPA is typically faster and much less prone to errors when compared to manual alternatives.

What is IoT?

IoT is mainly a collection of technologies that enables the remote measurement and, at times, control of physical elements. In recent years, cost reduction trends have made the required sensors, hardware, connectivity, and storage more affordable, expanding the reach of IoT to a broader audience and making it applicable to a wider range of uses.

A common example of IoT involves automating the measurement, logging, and reporting of important characteristics of physical objects. This could include monitoring chemical tank levels, HVAC operating parameters, equipment or vehicle locations, and determining the presence or absence of fully charged fire extinguishers in a high-rise building.

These are just a few examples of the many instances where data has traditionally been collected manually by humans. Although there is some automation in place, such as human operators using tablets or phone apps to record data, a large number of these tasks still require manual entry into back-office systems. This is often done by human operators who read from handwritten paper reports obtained from the field.

Most modern IoT systems do more than just collect data. They utilize a range of techniques, from simple algorithms to complex ML/AI models, to ensure data accuracy, cross-reference it with external information, analyze for irregularities, and even create new data from the collected information.

And just as RPA facilitates data exchange between systems, IoT enables the automation of manual “physical world” tasks, resulting in reduced human involvement, improved data accuracy, and faster data delivery. For instance, if a manual process is currently performed weekly due to travel time and resource availability, the IoT process could be carried out on a daily or hourly basis.

The potential of IoT extends beyond just automation. For businesses currently using or considering RPA, viewing IoT in this way could lead to new opportunities and similar benefits to those gained from RPA.

RPA or IoT? Or Both?

When embarking on a digital transformation initiative, it’s crucial to assess the current state of the business, pinpoint opportunities, and establish clear, measurable goals (often referred to as a “maturity assessment”) to ensure success. It’s essential to fully comprehend these elements before making investments in RPA or IoT.

The process of digital transformation involves incorporating various approaches and technologies. Companies can leverage both RPA and IoT to achieve their digital transformation goals. The role of each of these technologies may vary significantly in the initial stages of a digital strategy compared to later stages when the transformation becomes enterprise-wide.

A Proof of Concept (PoC) is usually the initial implementation of an IoT project. It often targets a specific use case that affects a particular process or area of a business. After a successful PoC, further projects can broaden the scope and impact on the business. RPA can also start as a PoC, with additional automation carried out as results are achieved.

Conclusion

The capabilities of modern RPA platforms include quick automation of specific business processes through core functionalities such as “bot creation,” rules processing, and screen scraping. However, they are not typically well-suited for IoT applications. In contrast, modern IoT platforms are specifically designed to interact with sensors, actuators, and other physical hardware, offering functions such as device and network management and data visualization that RPA platforms generally lack. Both platforms share some overlapping capabilities, such as API integration, automated communications, ML/AI support, and complex logic.

There is a strong argument for considering both RPA and IoT as part of any digital transformation journey, as stand-alone solutions, each with its strengths, and as combined and integrated solutions that leverage the strengths of both.

It is essential to note that both RPA and IoT can contribute to an overall digital transformation strategy. RPA is suitable for business processes that mainly handle data exchange between systems, while IoT is beneficial for business processes involving the physical world. Some processes may be addressed exclusively by either RPA or IoT, while others may benefit from the use of both.