Owen Molloy

A Framework to Support Distributed Business Process Analytics

Business Data Analytics is a powerful emerging technique to assist business users in decision making, providing end-users with visibility on process and business performance. There is an increasing demand for more advanced analytics for distributed business processes (e.g. supply chains), such as root cause analysis of performance issues, predictive analysis and the ability to perform ¡°what-if¡± type simulations. This requires a framework which supports business process modelling, real-time event capture, analysis and integration with advanced predictive analysis and simulation tools. This research project builds on our existing capability in monitoring business processes, and will focus on developing advanced cloud-based analytics capabilities for real©time analysis and optimization. This work will entail porting of recently developed tools ((Alejandro Vera Baquero and Owen Molloy ¡¯A Framework to Support Business Process Analytics¡¯, RDBPM 2012)) to a hybrid Distributed / Cloud-Based architecture and the incorporation of Big Data analytics capability. The project will also entail the incorporation of predictive analysis techniques and integration with simulation tools, leveraging and extending data exchange standards in these fields.

Distributed Risk Assessment

In many supply chains, risk is associated with both quality of product and continuity of service and / or production along the supply chain. Whether in the food and beverage industries, healthcare, manufacturing or other industries, they rely on the application of risk management on a continual basis to identify and quantify potential risk. The combination of multiple low-level risk can often manifest at multiple points in the supply chain and cause serious problems in terms of cost and liability. For example in the soft drinks industry, raw materials are combined at multiple stages with water, concentrates and other chemicals to produce products which are then subject to long and complex logistical and storage conditions, all of which are potential root causes of risk. Risk assessment techniques have evolved considerably in recent years to include modelling of causal mechanisms and application of Bayesian networks in predictive analysis.

This research project will bring investigate the application of the above techniques in tandem with data acquisition (building on our previous research in distributed business process monitoring) and analytics to create a framework and tools for modelling, monitoring and analysis of risk in distributed supply chains. We will work with our industry contacts in healthcare and the food and beverage industries to develop industry-based case studies and prototype solutions.

Collaborative Care Pathway Modelling and Management

The clinical data and systems environment in most public health jurisdictions can be characterized as a collection of multiple, fragmented, stand©alone applications and data sources. This fragmentation is a major impediment to the realization of integrated healthcare. Integrated care pathways provide a process view, based around patient journeys through a healthcare system. The introduction of care pathways delivers improved outcomes, efficiencies and more subjective benefits in terms of teamwork and patient satisfaction and confidence. Key to the successful implementation of process improvement initiatives are agreed process models and access to accurate performance data. Modelling tools are not currently available to support distributed teams to collaborate on pathway modelling.

This research will incorporate working with and extending existing modelling standards (e.g. BPMN, EHR) and leveraging online collaboration technology such as Apache Wave to develop techniques and tools to support collaborative modelling of care pathways by distributed multidisciplinary teams. We will extend current process modelling languages and tools to support domain specific modelling techniques for design, visualization and performance measurement of healthcare processes. This will also include semantically rich models of the roles, activities and information present in the process, as well as modelling of metrics, goals (e.g. performance targets, service level agreements) and constraints.