By Nilay Sharma, Founder – Director, Proteus Technologies Pvt. Ltd
As businesses seek to do more with their data, the integration of Artificial Intelligence (AI) into ERP has emerged as a game-changer across various verticals. In the age of digital transformation, Enterprise Resource Planning (ERP) systems has been the backbone of many organizations, helping streamline processes and improve operational efficiency.
The introduction of ChatGPT’s generative AI platform in November 2022 surprised ERP vendors and customers, leading to a re-evaluation of AI’s role in ERP systems.
Before, vendors offered AI development platforms, machine-learning chatbots, predictive analytics and specific applications for routine tasks.
ChatGPT’s human-like communication capabilities expanded the possibilities, enabling it to understand user intentions and handle more complex business processes when combined with mature machine-learning technology.
Experts now believe that generative AI has the potential to revolutionize ERP user interaction and automate a broader range of business operations than previously anticipated.
Here are some areas where significant progress has already been made:
Eliminating data entry – Using Generative AI
In the current usage of ERP system someone documents the transactions, sends it for approval, checks if it within the realm of the process defined and then it need to be “inputed” in the right manner in the ERP for further use and planning. This could be a Sales Order, Purchase Order, A payment to be made or an expense to be created, there has to be a documentation which needs to be interpreted, follow a process or an SOP and then if i may say “accounted for” in the ERP.
ERP solutions using AI now offer a method of getting this ‘accounted for’ using the source document. For example, the ERP could complete a sales transaction and all precceding processes by interpreting a mail and attached documents. For eg.
“Please raise a sales order to Acme distributors, Mumbai from the attached purchase order for A4 size paper of 80 GSM 100 quantity at a rate of Rs 100 for 500 nos. to be delivered within 10 days, Payment terms are 20 days from the date of dispatch. Provide a discount of 10% only for this order. This discount has to be charged to the “promotion” cost center. Please dispatch this only by Blue Dart courier”
This ERP will not only read the mail content but also the attached purchase order, interpret it, ensure all processes and SOP in the ERP, create a sales order, create a disptach through blue dart , create an invoice and account for the discount as suggested.
From the mail to the creation of the transaction is a matter of seconds!
This is not the future but such AI already exists. Currently, most organisations are” looping in” a human for final validation or approval, since this is new, we see that also going away and only exceptions detected by AI would need a human Looped in
This will do-away with weeks and months of training of the user interface and navigation of the system
Similar data and reports are also available using NLP; for example , “give me the outstanding for customer Ranvir singh, with segregation of over 60 days of due and mail it to the Client”
AI Copilots
AI co-pilots, also known as intelligent assistants, are making strides in ERP systems. These AI assistants offer suggestions and guidance to users, akin to a co-pilot in aviation. They assist in transaction creation, suggest next-best actions, and improve user experiences over time by learning and adapting to user preferences. As these AI co-pilots are capable of handling cross-module transactions, they are briging about an evolution in the user interface (UI) and user experience (UX).
Configuration of the ERP -a nightmare for most
Configuring an ERP system can be a daunting task, especially when businesses undergo changes and need to reconfigure the system. AI, coupled with machine learning, can simplify this process. Users can instruct the ERP to make specific configurations, which would traditionally take months, and AI can execute these changes within seconds. AI can also aid in generating test data and test cases, further reducing the complexity of system reconfiguration
For example just tell the ERP, “Create a new Profit Centre that will market a new product group called Wellness product group, which will have a separate sales team and a separate Invoicing series, These products will be sold only through our outlets in Mumbai Delhi and Ahmedabad.” and it make the configurations based on the interpretation of the text coupled with the learnings of the past”
Most of these AI-driven advancements are already available through ERP vendors.
ERP vendors like us have just begun exploring how we can leverage AI. The speed at which AI is advancing vendors like us are rethinking the entire process and using the ERP only as the broker and transaction layer, not necessarily as the system users interact with.
On the lighter side I won’t be surprised if someone comes out with a AI plugin that even suggest that what you want to achieve can be achieved through another solution and not the one that you are using!!
