No abstract

No abstract

Information systems leave a traceable digital footprint whenever an action is executed. Business process modelers capture these digital traces to understand the behavior of a system, and to extract actual run-time models of those business processes. Despite the omnipresence of such traces, most organizations face substantial differences between the process specifications and the actual run-time behavior. Analyzing and implementing the results of systems that model business processes tend, however, to be difficult due to the inherent complexity of the models. Moreover, the observed reality in the form of lower-level real-time events, as recorded in event logs, is seldom solely explainable by higher-level process models. In this paper, we propose an architecture to model systemwide behavior by combining process mining with a multi-agent system. Digital traces, in the form of event logs, are used to iteratively mine process models from which agents can learn. The approach is initially applied to a case study of a simplified job-shop factory in which automated guided vehicles (AGVs) carry out transportation tasks. Numerical experiments show that the workflow of a process mining model can be used to enhance the agent-based system, particularly, in analyzing bottlenecks and improving decision-making.

In order to remain competitive, logistics companies are forced to provide smart solutions within a network that is characterized by complexity and heterogeneity. The advancements of sensing and communication technologies stimulate logistics organizations to improve their business performances by using more advanced decision support tools. This research is devoted to improve logistics decision making by exploiting the enormous datasets originating from IoT networks in combination with Big Data Analytics. The main aim is to develop a resilient planning framework that stimulates logistics planners to combine both human experiences and pattern recognition mechanisms (e.g., machine learning, data mining, etc.). In this paper, four research deliverables are proposed to pursue this vision: (1) a state-of-the-art overview of modern decision support tools to enhance logistics resilience and efficiency; (2) the development of dynamic optimization algorithms using real-time data; (3) the construction of data-driven algorithms to identify, assess and resolve the presence of logistical disturbances and; (4) the formulation of resilient planning framework that enables real-life implementations of the algorithms developed. A brief overview of the required research activities is given as well, including a visualization of the activities' coherency. This paper concludes with a description of the preliminary results and some future research directions.

Increasing customer demands and variability in today's logistics networks force fleet operators to become more reliable and flexible in their operations. As modern-day fleets are well equipped with wireless sensing, processing, and communication devices, fleet operators could proactively respond to dynamic events. However, the use of real-time sensor data to achieve re-optimization is scarce. This observation raises the question of how logistics operators should incorporate the emerging track-and-trace services into their dynamic planning activities. In this paper, we propose a reference architecture that relies on both the Internet of Things and the Smart Logistics paradigms, and aims at enhancing the resilience of logistics networks. Since the decision of when to reschedule the network's configurations remains nontrivial, we propose a hierarchical set of disruption handling systems to facilitate the trade-off between decision quality and response time. In our design, autonomous logistics agents can quickly anticipate on minor changes in their surroundings, while more severe disruptions require both more data and computational power in higher-level processing nodes (e.g., fog/cloud computing, machine learning, optimization algorithms). We illustrate the need of our architecture in the context of the dynamic vehicle routing problem.