Based on the recent trends shaping digital transformation, business leaders agree they must significantly accelerate the implementation of digital technologies such as robotic process automation (RPA) and intelligent automation (IA) powered by computer vision, natural language processing (NLP), machine learning (AI-ML), and low code no code (LCNC) platforms to effectively support the business. When implementing RPA, CXOs see investment in process mining and task mining as key to unlocking returns from the technology.
Despite ongoing investments in RPA, CXOs feel the need for hyper automation to fully realize their automation objectives. To realize higher value from their RPA investments, they are turning to a suite of data mining technologies, such as process mining and task mining which are future drivers of growth for automation using RPA and IPA in the coming years.
Process mining allows business units to gain complete transparency into business processes to transform them with automation and improvements that drive continuous operational efficiency. Digital footprints left behind in ERP, CRM, and PLM systems have a story to tell about how work really gets done, which process transformation actions are most impactful, and where your best automation opportunities lie. It provides sharper insights to continuously optimize, automate, and monitor end-to-end processes.
As people interact with business IT systems, their actions are captured using the digital footprints left by human users. This data can be transformed into pre-formatted event logs and visualized with the help of a process mining tool. Process mining comprises of four stages—collection, discovery, enhancement, and monitoring and conformance checking.
Involves collection of time-stamped event log data from key transactional systems such as SAP, Salesforce, Oracle, PLM, etc. The activity or interaction of humans with the system creates a digital record. Receiving a quote, creating a purchase order, submitting documentation for loan approval, approving a loan, entering information into a health EMR system are examples of activities that can be mined for data. Involves collection of time-stamped event log data from key transactional systems such as SAP, Salesforce, Oracle, PLM, etc. The activity or interaction of humans with the system creates a digital record. Receiving a quote, creating a purchase order, submitting documentation for loan approval, approving a loan, entering information into a health EMR system are examples of activities that can be mined for data.
This stage involves discovery of real processes as they actually happened within the log file data. Process mining software transforms the digital records (log file data) into event logs. The most common format for these event logs is an XML-based format called XES (extensible event stream) which was adopted by the IEEE Task Force on Process mining. Event logs have at least three main attributes: case ID, activity, and timestamp.
This involves enhancement of the process mining scope to optimize business outcomes. The enhancement can be a process re-design or reengineering by eliminating duplicates, in effective process steps to eliminate rework, increase efficiencies, and for automating the manual rules-based steps using RPA. The visualization of a process is automatically created using event logs. It is important to understand that unlike traditional business process management (BPM) techniques, process mining shows the real process based on evidence residing in the log file data. The process mining tool can mine data of the last six months to six years depending upon the requirement and scope agreed upon with the process owner.
This stage involves monitoring of the process improvement changes that were made in Stage 3 for further improvement opportunities. Here, KPIs can be created and monitored to uncover potential improvement areas; data mining and/or ML algorithms can be used to detect hidden patterns and dependencies; or conformance checking techniques can be applied to compare the process to a certain ideal model. Conformance checking compares the actual process with a predefined model to discover deviations. So, it is used to check if the reality conforms to an existing pattern. Conformance checking is very useful to audit and for compliance teams to identify non-compliance and deviations from standard operating procedures.
Task mining is a technique to uncover business processes by recording user interactions with various systems, including enterprise solutions (ERP, CRM, BPM, ECM, etc.), personal productivity applications (Microsoft Excel, Outlook, Word, PowerPoint etc.), and terminal and virtual environments. Unlike process mining that reconstructs processes using the data from event logs generated by enterprise systems, task mining captures user interactions with any application. It records necessary data such as keyboard actions, mouse movements, clicks, etc. for identifying RPA opportunities and automatic creation of process discovery documents (PDD) and RPA bots.
Process and task mining help identify inefficiencies, bottlenecks, and areas of improvement in business processes. By analyzing the captured data, organizations can gain insights into how tasks are performed, uncover deviations from the intended process, and optimize workflows to enhance efficiency and productivity.
Process and task mining can assist in compliance monitoring and risk management by capturing and analyzing data related to regulatory processes. They enable organizations to identify compliance gaps, detect potential risks, and ensure adherence to regulations and internal policies.
Process and task mining provides insights into how employees perform tasks, allowing organizations to assess individual and team productivity. By understanding how tasks are completed, organizations can identify training needs, develop targeted training programs, and provide personalized coaching to improve employee performance.
Process and task mining data can be used to identify suitable processes for automation. By analyzing repetitive and rule-based tasks, organizations can prioritize automation efforts, streamline processes, and leverage robotic process automation technologies to automate tasks, freeing up human resources for higher-value activities.
Process and task mining can help organizations understand the customer journey by capturing and analyzing user interactions with digital platforms. By analyzing these interactions, organizations can identify pain points, optimize user interfaces, and enhance the overall customer experience.
Process and task mining can be applied to IT operations to monitor and optimize system performance. They can capture and analyze data related to software usage, system configurations, and user interactions, enabling organizations to identify performance issues, improve system efficiency, and make informed decisions regarding IT infrastructure.
Process and task mining can be used as tools to identify potentially fraudulent activities within an organization. By analyzing user interactions and transactional data, organizations can detect anomalies, patterns, and suspicious behaviors, helping to prevent and mitigate fraud.
These are a but a few examples of the applications of process mining and task mining. The technology has the potential to provide valuable insights and drive improvements across various aspects of an organization's operations.
Identifying and understanding hidden inefficiencies in end-to-end processes is an important step toward a resilient and successful business. But small manual tasks are often overlooked when it comes to process improvements and process automation initiatives. This, however, can lead to long waiting times for customers, inefficient handovers, underutilized resources, non-compliant procedures, and failure of RPA and digitalization efforts.
Thus, combining process mining capabilities with task mining provides a comprehensive picture of core enterprise processes. It removes process gaps and eliminates hidden inefficiencies and noncompliance risks. It allows process owners to achieve full transparency with their processes, bridging the gap between task analysis and end-to-end process view.
Prakash Narayanan is Solutions Head for RPA and intelligent automation solutions at Cyient. He has over 24 years of experience in IT and has delivered 1000+ bots across the banking, pharmaceuticals, telecom sectors, and has extensive experience in Intelligent process automation tools and platforms. He was named among the Top 16 Global Automation Rockstars by Dynamic CIO.
