Where’s the ‘Coleman’ for IBM i?
August 30, 2017 Alex Woodie
In case you haven’t noticed, we’re in the midst of an artificial intelligence (AI) revolution that’s bringing automation to places where it never existed, including ERP software. The latest entry into the field is Infor, which turned heads last month with a new AI bot dubbed Coleman. Unfortunately – or fortunately, depending on your views on AI – Coleman won’t be coming to IBM i.
Infor, which owns the biggest collection of IBM i application software on the planet, has slyly positioned itself as a scrappy alternative to Oracle and SAP in the enterprise software space. While the giants from Germany and California cater to the world’s biggest firms, Infor smartly lays claim to the vast midmarket, where the majority of its 18,000 or so IBM i customers play.
But that midrange focus doesn’t mean less sophisticated software. In fact, with its new Coleman AI offering, one could make the argument that Infor is pushing the envelope in terms of what ERP software can do.
If you’ve ever used Apple‘s iPhone chatbot Siri, then you have an idea of what Coleman can do. The software, which is based on the same framework that Amazon Web Services uses for Alexa, can answer simple questions posed by users, including “What’s my PTO balance?” by automatically generating a SQL query and then presenting the results through a Birst BI report (Infor just bought Birst, you will recall).
Coleman can also do more advanced things, including automating repetitive tasks like invoice matching, or recommending a next-best actions, similar to how Netflix makes movie recommendations or Pandora makes song recommendations.
Eventually, Coleman – which was named after Katherine Coleman Johnson, the NASA mathematician who calculated orbits during the early days of the space race and whose life was featured in the 2016 film “Hidden Figures” – will be called on to tackle what Infor co-president Duncan Angove called “low-complexity, high-volume” tasks, such as expense report reconciliation, journal reconciliation, evaluation of credit risk for new customers, reconciling deductions in AR, performing inter-company transactions, creating consolidated reports, and closing the books.
“With Coleman, we seek to maximize human potential,” Angove said during last month’s Coleman launch event at the Inforum conference in New York City. “What if you could make the best decision every time? Do the work of two people and be relieved of repetitive transactional tasks that allow you to spend more time with customers or patents or citizens?”
Angove cited research stating that 40 percent of the work of finance departments will eventually be automated. “AI is the new steam engine,” he said. “It’s the new electricity and it will transform everything. Society is changing one learning algorithm at a time.”
Coleman leverages the familiar technological stack that Infor has been building for some time, including the ION data integration bus and its Ming.LE interface. These technologies are available to Infor’s IBM i customers who are using the latest releases of the IBM i-based ERP systems that have been adapted to run with the Infor Xi (previously 10x) platform.
Unfortunately for Infor’s IBM i customers, Coleman won’t be coming to Infor’s IBM i products. The software requires the ERP suite to be running on the AWS cloud, and since AWS only offers X86 servers in its cloud, IBM i applications cannot run there. “Due to the Amazon technologies Infor leverages for Coleman, including Amazon Lex for conversational UX, it is not able to offer Coleman on IBM i,” an Infor spokesperson told IT Jungle.
It’s too bad that the thousands of companies relying on IBM i-based ERP packages like Infor ERP LX (BPCS), ERP XA (MAPICS), ERP System21, and ERP M3 to run their businesses can’t take advantage of Coleman to automate tasks. These ERP systems were once at the cutting edge of technological sophistication, and while they may not have the most brilliant user interfaces compared to the standards of the day, they still offer core business functionality that’s as good, if not better, than comparable products from SAP, Oracle, Microsoft, and SAGE.
The good news is it’s not a matter of “if” Coleman (or something like it) comes to IBM i, it’s just a matter of “when.” The forces being unleashed by AI technology acting upon big data sets are too great to ignore – even too great for normally conservative ERP software developers to ignore.
The latest wave of innovation has echoes of the past. Just as having a centralized suite of business applications built upon a common database (i.e. the modern ERP suite) displaced millions of employees, AI-based tasks that have been trained with machine learning algorithms to simulate human actions will eventually eliminate the need to have a human behind the mouse and keyboard.
The need for something like an ERP system won’t go away. But the way that the ERP system is built and the way it’s designed to interact with humans will definitely change. The first vendor that can figure out how to leverage AI technology to essentially drive ERP software without the need for so many human workers will have a leg up on the competition.
It’s something that CRM pioneer Tom Siebel, who now heads up the IoT data analytics company C3 IoT, understood when he predicted recently that AI is “a 100 percent replacement market” for ERP, supply chain, and manufacturing software. It’s something that Lane Nelson, who is president of an IBM i-based ERP software company called HarrisData that’s researching AI’s potential in business software, understands. It’s something that’s understood by Rocket Software‘s Dan Magid, who advises IBM i clients to consider next-generation user experiences as central to their business transformation strategy.
In the end, Colemans will become commonplace in our future world. If a chatbot can deliver an account balance or recommend a product for a customer as well as a human can, then it’s a just a matter of time before that money-saving tactic becomes the norm.
The key question is how soon this sort of technology becomes commonplace on the IBM i platform. Much of the emerging and data analytic fields are fully committed to the open source software development and distribution method, so the availability of the underlying technologies shouldn’t be a problem. Getting them to run on IBM i and hammering them into a workable product, however, will take time.