What are the best uses for RPA and AI in ERP?
AI and robotic process automation are taking over more and more tasks from ERP users. RPA mimics human behavior and records users as they enter data, execute commands, and move documents between applications.
The machine learning AI in ERP scans information for patterns, “learns” what to expect, makes decisions and even tries to predict the future.
Both technologies enable a new breed of ERP that is more automated, responsive, and easier to use than their often frustratingly headstrong ancestors.
Bob De Caux deals with such questions in his role as Vice President of AI and RPA at IFS, the Swedish ERP provider. Together with myself and Brian McKenna, Business Applications Editor at ComputerWeekly, he had an extensive discussion about the respective roles of RPA and AI in ERP and the evolution of the technology.
The use of AI in ERP has increased in the last five years and with it the frequency of AI buzzwords that are applied to functions that are more owed to traditional IT with its binary logic and decision trees.
It would help define the terms – intelligence to start with.
“When we think of hard-coded rules, they are set in stone, they don’t change, they don’t adapt,” said De Caux. Instead, to be considered intelligent, software must be able to “adapt and learn and find deep complex patterns in the data itself without having to be hard-coded”.
Bob De Caux
Intelligence is also manifested in how software represents knowledge, similar to how the brain works. “It could be more complex hierarchies, it could be relationships with different strengths,” he said.
It doesn’t help that vendors use the term AI too liberally in areas like healthcare, while conventional data analytics and IT are actually driving much of the advances. McKenna discussed the problem in a recent blog post on Washing AI.
De Caux agreed that exaggerating the term would not help. “It has raised expectations of what AI can deliver, and certainly what it can do without a lot of setup work.”
‘Robotics’ is correct in the name
RPA, on the other hand, often uses AI, but is not itself a type of AI. “It tries to effectively replicate how a person interacts with a system, in terms of the clicks they make and the processes they follow.” This involves solving many complex problems, using AI in things like computer vision, understanding buttons that appear on a screen, and dealing with uncertainties about users’ intentions.
“There are many uses for AI within RPA, but does it solve a problem that artificial intelligence should really be applied to? Not really,” he said. “There is a real attempt to automate a workflow,” which can often be done more effectively with systems integration tools like APIs.
RPA is a great way to replicate what a person does, but it doesn’t scale or adapt well to change and can result in an inefficient system being replicated. It often makes more sense to use machine learning to automate and optimize the workflow.
De Caux said many of the early enterprise artificial intelligence applications were geared towards extracting value from data, a topic he discussed in a post on the IFS blog. But it can’t just be any data. The data set must be large enough to train intelligent algorithms and have patterns that the algorithms can identify, he wrote.
The next level will be intelligent process automation that solves more complex problems than the automation of repetitive, time-consuming tasks that is at the heart of the current generation of RPA and AI in ERP. De Caux compares this emerging intelligent automation of business processes in the System of Record – ERP – to the intelligence in industrial automation.
De Caux also stated:
- the tasks for which AI is best suited;
- where it fits into key ERP functions like human capital management and CRM;
- the importance of “explainable” AI; and
- What’s next from IFS when using AI in ERP.
To hear the discussion, click the podcast link above.