Data Analytics for Process-Oriented Applications

October 12th, 2020

 The Challenge

Business processes are an integral means through which individual organisations or networks of organizations such as supply chains deliver services to their customers. Business Process Management (BPM) is dedicated to the development of methods, techniques, and tools for (re-)designing, implementing, automating, monitoring, and analyzing business processes, in order to deliver efficient and effective business processes.

Our Response

Our BPM research focuses on the development of approaches through which processes can be analyzed and automated. To this end, we address a variety of problems relevant to both,  the business as well as the software development perspective.

  • Data-driven process analytics: We develop techniques for analysing event data that is created when processes are executed. We support analysts in mining such data and in validating analysis results obtained from such data, ultimately enabling organizations to derive actionable insights such as bottlenecks and optimization opportunities. In addition, we devise techniques tailored towards specific application scenarios such as the analysis of blockchain applications or DevOps processes.
  • Process execution and enforcement on blockchain: We designed approaches to model-driven development of smart contracts. Instead of manually implementing smart contracts, using our approach developers specify applications at a more abstract level, focusing on the business processes and the data that are handled by the application. The specifications are then automatically translated into highly optimized source code that can readily be deployed on the blockchain.
  • Automated analysis of process documentation: Process models, albeit created manually or discovered from event data, capture knowledge about business processes and constitute a valuable information source for many tasks in the BPM lifecycle. We developed techniques that assist experts in navigating collections of such models along implicit functional relationships by automatically identifying functional replications and consolidation opportunities.

The Results

Our research has resulted in a set of tools that can readily applicable tools and frameworks:

  • Ethereum Logging Framework (ELF): is an open-source framework that provides advanced logging capabilities for Ethereum applications. It consists of four components. Ethql is a query language for retrieving data stored on Ethereum blockchains, the validator checks ethql scripts for specification erros, the extractor retrieves, transforms and formats data based on ethql scripts, and the generator creates efficient logging functionality that can be embedded into smart contracts.
  • Lorikeet: A development environment for executing (collaborative) business processes and hosting data registries on the blockchain. Lorikeet adopts a model-driven engineering approach. That is, developers can focus on specifying the processes and registries as models, while Lorikeet automates the deployment of these models on the blockchain by generating, compiling and deploying the source code.
  • Process-Oriented Dependability (POD): The POD framework aims to improve the dependability of processes through behavioral anomaly detection. The approach relies on creating a process model for each operation and using that model to guide near real-time detection, diagnosis, and recovery from errors in the execution of the process.

To find out more about our research, contact Christopher Klinkmueller (christopher.klinkmueller@data61.csiro.au) or browse our publications:

  • C. Klinkmüller & I Weber (2021): Every apprentice needs a master: Feedback-based effectiveness improvements for process model matching. In: Information Systems 95, 101612.
  • C. Klinkmüller, I. Weber, A. Ponomarev, A.B. Tran & W. van der Aalst (2020): Efficient Logging for Blockchain Applications. arXiv:2001.10281
  • C. Klinkmüller, R Müller, I Weber (2019): Mining Process Mining Practices: An Exploratory Characterization of Information Needs in Process Analytics
  • C. Klinkmüller, A. Ponomarev, A.B. Tran, I. Weber & W. van der Aalst (2019): Mining blockchain processes: extracting process mining data from blockchain applications. In: International Conference on Business Process Management (Blockchain Forum Track).
  • C. Klinkmüller, N.R.T.P. van Beest & I. Weber (2018): “Towards reliable predictive process monitoring”. In: International Conference on Advanced Information Systems Engineering (Forum Track).
  • C. Klinkmüller & I. Weber (2017): “Analyzing control flow information to improve the effectiveness of process model matching techniques”. In: Decision Support Systems 100, pp. 6-14.