As an expert in decision support systems development, I have been promoting transparency and self-explanatory systems to close the plan-do-check-act cycle. AI adoption has tripled in 2018, moving AI towards the Gartner-hype-cycle peak. As AI is getting more mainstream, more conservative companies have good reasons to enter this arena. My impression is that the journey is starting all over again as organizations start using AI technology as black box systems. I think that eventually these companies will also start asking for methods that result in more reliable project outcomes and integrated business systems. The idea that explainable AI is at the expense of accuracy is deeply rooted in the AI community. Unfortunately, this has hampered research into good explainable models and indicates that the human factor in AI is underestimated. Driven by governments asking for transparency, a public asking for unbiased decision making and compliance requirements on business and industry, a new trend and research area is emerging. This talk will explain why explainable artificial intelligence is needed, what makes an explanation good and how ontologies, rule-based systems and knowledge representation may contribute to the research area named XAI.
Many organizations employ an information system that supports the execution of their business processes. During the execution of these processes, event data are stored in the databases that support the information system. The field of process mining aims to transform such data into actionable insights, which allow business owners to improve their daily operations. For example, a process model describing the actual execution of the process can be easily extracted from the captured event data. Most process mining techniques are "backward-looking" providing compliance and performance information. Few process mining techniques are "forward-looking". Therefore, in this paper, we propose a novel scenario-based predictive approach that allows us to assess and predict future behavior in business processes. In particular, we propose to use system dynamics to allow for "what-if" questions. We create a system dynamics model using variables trained on the basis of the past behavior of the process, as captured in the event log. This model is used to explore the effect of possibly applied changes in the process as well as roles of external factors, e.g., human behavior. Using real event data, we demonstrate the feasibility of our approach to predict possible consequences of future decisions and policies.
We often find it challenging to translate a legacy system when the software business is critical. Adding to the misery of technical debt is the “Broken Window” concept which adds more complexity in exercising dynamic context resolutions for independent services alongside governance and data management. This often leads to a maze of disoriented services with high interdependency. To seriously adopt “Operate what you Build” phenomena, we need a granular facade approach to understand the business requirement and translate it to the architectural operators. The paper tries to provide an approach to establish platform independent interfaces, bounded domain contexts, eliminating non-critical legacy components and incremental quality aware methods to translate a legacy system to microservices. Along with the architectural objects, the paper will also present granular level of performance management of the translated application to consider factors like - system, container, network and application service itself.
This paper proposes a novel crossing-workflow fragment discovery mechanism, where an activity knowledge graph (AKG) is constructed to capture partial-ordering relations between activities in scientific workflows, and parent-child relations specified upon sub-workflows and their corresponding activities. The biterm topic model is adopted to generate topics and quantify the semantic relevance of activities and sub-workflows. Given a requirement specified in terms of a workflow template, individual candidate activities or sub-workflows are discovered leveraging their semantic relevance and text description in short documents. Candidate fragments are generated through exploring the relations in AKG specified upon candidate activities or sub-workflows, and these fragments are evaluated through balancing their structural and semantic similarities. Evaluation results demonstrate that this technique is accurate and efficient on discovering and recommending appropriate crossing-workflow fragments in comparison with the state of art's techniques.
The paper presents an Integrated Maintenance Decision Making Model (IMDMM) concept for cranes under operation with dependent stochastic function into the container type terminals. The target is to improve cranes operational efficiency through minimizing the risk of the Gantry Cranes Inefficiency (GCI) results based on implementation of copula approach model for stochastic degradation function dependency between cranes. In the present study, we investigate the influence of dependent stochastic degradation of multiple cranes on the optimal maintenance decisions. We use copula to model the dependent stochastic degradation of components and we formulate the optimal decision problem based on the minimum GCI expected. We illustrate the developed probabilistic analysis approach and the influence of the dependency of the stochastic degradation on the preferred decisions through numerical examples, and we discuss the close relationship of this approach with interoperability concepts. The crane operation risk is estimated with a sequential Monte Carlo Markov Chain (MCMC) simulation model and the optimization model behind of IMDMM is supported through the Particle Swarm Optimization (PSO) algorithms.
