Enterprises have long relied on business process management to support their digital transformation and process engineering initiatives. AI is now providing a significant boost to BPM.
According to Jeff Springer, principal consultant at data and analytics firm DAS42, “AI technology is rapidly advancing, making it possible to develop more sophisticated and effective AI-powered process discovery and automation solutions.” He continued, “Many of these advancements are due to the increasing availability of data from many sources, including enterprise systems, sensors, and social media, which leads to larger-scale AI deployments.” For instance, AI systems can now learn from data and recognize patterns that would be challenging or impossible for humans to recognize thanks to the development of deep learning algorithms.
In what ways is AI changing BPM?
AI-enabled deployments are finding a wide range of uses in business process management (BPM), from enhancing front-office operations to leveraging generative AI process modeling capabilities to mapping business processes.
Front-desk procedures
According to Brian Steele, vice president of product management at call center intelligence platform provider Gryphon, AI deployments in front-office processes are boosting sales, raising customer satisfaction, and enhancing employee engagement. AI-powered business process management, for instance, is enhancing customer interactions in contact centers by reducing call wait times, personalizing recommendations, and offering real-time sales support.
Mining processes
Process mining is a crucial BPM enabler that aids companies in finding ways to enhance operations, add value, and cut expenses. According to Chris Monkman, vice president of product management, AI, and knowledge at business process software provider Celonis, “process mining makes the data that the AI [system] is being trained on much more intelligent, unlocking its true power.” In contrast, process mining makes AI much faster and easier to use. However, advancements in real-time structured data and semantic knowledge will be necessary for training large language models (LLMs) and generative AI’s battle with hallucinations.
Mining processes with an object-centric approach
To better understand and manage business processes, Celonis and RWTH Aachen University are combining AI with object-centric process mining, which represents actual objects and events in a process. AI is able to send alerts in the event of delays, update expected delivery times continuously, and even take remedial action when a real object—like a shipping order or invoice—moves through the business process.
Big process models
The process management startup SAP Signavio trains what it refers to as large process models (LPMs) to analyze process data more precisely by using labeled data in LLMs. The SAP Signavio Academic Models LPM data set, which comprises hundreds of thousands of business models primarily in Business Process Modeling Notation, was made available by SAP and academic researchers. According to Dee Houchen, head of global market impact at SAP Signavio, LPMs could be used in a variety of use cases, including process analytics, content creation, best practice recommendations, and process data augmentation.
Extraction and enrichment of data
According to Bruce Orcutt, senior vice president of product marketing at ABBYY, the company that makes optical character recognition software, it is investigating how AI technologies can extract more data from customer documents and correspondence to speed up decision-making in enrollment, funding, and approval processes. AI may also be utilized to enhance process results and data insights. “Data is king,” Orcutt stated, “but AI helps to make sense and bring context and meaning to all the data in an impactful way to the business.”
Low- or no-code programming
Traditionally, BPM analysis tools are used in conjunction with low-code and no-code tools to help expedite business reengineering initiatives. With GitHub Copilot’s capabilities, AI is making it possible to develop more low-code or no-code applications, according to John King, partner in business processes at Lotis Blue Consulting. In order to meet customer needs, this capability promises more A/B testing types of deployments and faster change velocity, which can encourage the decentralization of application development. Companies can also create and maintain apps that automate crucial business procedures using just the platform and infrastructure support provided by IT.
Analysis of work networks
Graph theory is used in network analysis to comprehend the composition and operation of complex systems. King reasoned that by using work network analysis to process meeting, phone, email, and instant message artifacts, these same ideas might be applied to businesses. When productivity needs to be increased, AI can recognize and contrast patterns of cooperation and behavior with the standards and best practices of the organization.
Twin digital data
Digital twins are functional representations of complex processes and physical environments that are connected to the real world via a digital thread. AI can assist in converting unprocessed sensor and workflow data into more useful digital twins. According to King, AI can also be used with these models to provide various scenarios and decision analysis. “This will save time and money,” he said, “and allow firms to model rare aor prospective events before they occur, understand the impact of the event in a safe but also objective environment, and allow contingencies to be developed ahead of time.”
Mapping business processes
According to Springer of DAS42, AI and machine learning models are already being used to automatically map out business processes and find areas for automation and improvement. He mentioned a manufacturing company that has seen a 10% increase in production output as a result of implementing an AI system to monitor its production line in real time, identify potential bottlenecks and other issues, and recommend corrective actions to operators.
Analyzing business processes
Process experts have traditionally carried out business process analysis by hand. According to Stephen Ross, head of business development, Americas, at cybersecurity consultancy S-RM, artificial intelligence (AI) in business process management (BPM) could expedite the results of business process analysis for tasks involving modeling, collaboration, process mining, risk management, and compliance.
NLP, chatbots, and virtual assistants
While chatbots and virtual assistants have been around for almost 60 years in one form or another, only in the last ten years has their true business value been recognized. Natural language processing (NLP), which is powered by generative AI, creates new business opportunities for chatbots and virtual assistants that are integrated into BPM systems to resolve queries, assist staff members with tasks, and enhance customer interactions. Additionally, NLP excels at deriving insightful conclusions from the analysis of unstructured data sources like social media posts and customer reviews.