Session: CIE-05-01CIE Graduate Student Poster Symposium

Paper Number: 74730

Start Time: August 18, 10:00 AM

74730 - Exploring Machine Learning for Business Process Knowledge Extraction and Management 

Business process management is a critical activity that aims at increasing the efficiency of business operations. The global manufacturing industry is currently transforming smart manufacturing. Smart manufacturing is the synthesis of advanced manufacturing capabilities and digital technologies to collaborate and create highly customizable products. In a smart manufacturing environment, the requirement for products is more diverse, customized and complex, which also requires more effective business process management. Business processes management is a multifaceted problem that is comprised of analyzing both activities’ workflow, as well as the decisions that are made throughout that workflow. In process mining, the automated discovery of process models from event data, a strong emphasis can be found towards discovering this workflow and how data influences that workflow, i.e., decision point analysis. However, many process mining methods combining machine learning are rarely used in industry. Several challenges do exit if intelligent process mining methods are expected to be used in industry. For example, these methods are designed to be applied to original event logs. Because these methods do not consider other perspectives on the data that could be obtained by using data transformations, many patterns are missed that may represent critical information. To fulfil this gap, we aim to outline a process knowledge extraction and management framework that can support business process management and decision. The framework starts from multi-structured process data such as event logs, text and image. These process models and social networks are then extracted from these process data. The objective that we propose in this framework is to facilitate decision work and provide support for future business processes. Process models should focus on things related to specific types of users. The discovered model can focus on different aspects (control flow, data flow, time, resources, cost, etc.) and express it with different granularity and accuracy. Build such a model can provide different levels of processes for different participants, making the process model a continuous model. As a promising study, we believe that this research will significantly help to increase the efficiency of business operations.

Presenting Author: JUNYA TANG Tongji University

Authors: