Till Becker

Applying autonomous control methods in different logistic processes - A comparison by using an Autonomous Control Application Matrix

Autonomously controlled processes represent a decentralized approach to deal with todays increasing complexity in production logistics. Different autonomous control methods offer the possibility to redesign logistics processes in the way that control is shifted from centralized planning to autonomous logistic objects. The performance of these methods embedded in logistic processes needs to be measured in order to select the appropriate method for each process. This measurement can be done using an evaluation system but due to the vast diversity of logistic processes, it is extremely difficult to give general recommendations regarding the performance of autonomous control methods in different settings. Especially companies have to be able to find the most suitable methods for their individual logistic processes. This paper introduces an Autonomous Control Application Matrix which allows identifying the autonomous control method that fits best to a specific logistic process. The matrix contains actual performance values, performance percentage, and rank of each method in combination with a specific process.

Application of a three-component evaluation system for autonomous control in logistics

Abstract Autonomous control of logistic systems is a potential solution to the problems encountered in modern competitor marketplaces. Autonomous control in a logistic system is characterized by the ability of logistic objects (e.g. order, pallet, machine) to process information in order to render and execute decisions on their own. In order to determine the limitations of autonomy an evaluation system is needed which is presented in this paper. The evaluation system is created by three components which can be combined to analyse the potential of autonomous control. This potential is visualized by the result of the evaluation procedure with the help of a three-dimensional (3D) diagram built from single curves. Simulation studies are used to verify the surface of the 3D diagram and highlight the limitations of the autonomous control approach. All the varied parameters (level of autonomous control, level of complexity, and level of logistic objective achievement) are described and measured in the simulation study.

A Comparison of Network Characteristics in Metabolic and Manufacturing Systems

Both metabolic and manufacturing systems face fluctuating environmental influences and thus share the common challenge to maintain a high level of efficiency for a variety of different conditions. Therefore, transferring methods used for analyzing one of the systems can lead to gaining new insights in the other. Following-up on previous findings on analogies in metabolic and manufacturing systems, our approach now is to analyze and compare complex network measures such as centrality or flow activity in both systems to identify quantified relations. The results show that both systems also display distinct statistical differences in addition to their various structural similarities.

Process Maintenance of Heterogeneous Logistic Systems—A Process Mining Approach

Processes in manufacturing and logistics are characterized by a high frequency of changes and fluctuations, caused by the high number of participants in logistic processes. The heterogeneous landscape of data formats for information storage further complicates efforts to automatically extract process models from this data with the tools from Process Mining. This article introduces a concept for constantly updating process models in logistics, called Process Maintenance, collects requirements for a common view on information in logistics, and shows that Process Mining with logistic data is possible, but still needs improvement to become a regular practice.

Network structures and decentralized control in logistics: topology, interfaces, and dynamics

Decentralized or autonomous control in logistics has continuously developed over the last decades. Trends such as Ubiquitous Computing, Industry 4.0, and Cyber–physical Systems foster the transition from centralized control to decentralized approaches. While there has been a lot of research on control algorithms and miniaturization of information and communication technology, comparably little is known about the effect of the underlying structures of the logistic networks on decentralized control. This article aims at revisiting the topic of decentralized control with regard to the topology of logistic networks and at highlighting the importance of the relation between topological structure and control procedures in order to propose directions for future research.

A generic implementation approach of autonomous control methods in production logistics

Autonomous control in logistics is a decentralized approach for coping with rising complexity and dynamics in todays production networks. It is put into practice by redesigning logistics processes in the way that control is shifted from a centralized planning entity to autonomous logistic objects. Autonomous control methods are the algorithms that define how decisions are made by each individual object during production control. Although there are contradictory logistic targets as well as many different ways how decisions can be taken, it is possible to have a single implementation scheme that can be configured easily to cover many different variants of autonomous control methods. This paper presents a generic implementation approach that recursively collects any desired information used as basis for decision-making. Because it is transparent and easy to adapt, it can support both science and industry in developing, testing and implementing autonomous control.

Interoperability of Logistics Artifacts: An Approach for Information Exchange Through Transformation Mechanisms

The information flow in logistics is covered by a growing set of systems and standards. To handle the increasing heterogeneity, the exchange of information rather than the exchange of data is favored because unlike data information contains a precise meaning. This precise meaning would reduce the false interpretation between stakeholders within information flows. The exchange of information on the system level requires the application of transformations on the communication interface levels. In this paper, a corresponding transformation and communication approach is presented. Finally, the impact of the conceptual approach is shown on a hypothetical application scenario.

Applying Process Mining in Manufacturing and Logistic for Large Transaction Data

Process mining is a promising approach to extract actual business processes form event logs. However, process mining algorithms often result in unstructured and unclear process models. Moreover, sufficient data quality is required for accurate interpretation. Therefore, adopting process mining for the field of manufacturing and logistics should take into account the complexity and dynamics as well as the heterogeneous data sources and the quality of event data. Therefore, the objective of this work is to study the application of process mining in the manufacturing and logistics domain with real data from manufacturing companies. We propose a methodology to improve the limitations of process mining by using a Markov chain as a sequence clustering technique in the data preprocessing step and apply heuristic mining to extract the business process models. Finally, we provide results from an experiment with real-world data in which we successfully improve the quality of discovered process model in the regards of replay fitness dimension.

Social Context-Aware Recommendation for Personalized Online Learning

Abstract The integration of ICT in teaching and learning enables the paradigm shift for education system by creating a possibility for learner to learn anywhere and anytime through variety of communication system. To enhance effective learning for a large number of learners, online learning requires effective personalized learning method. For decades, recommendation system is responsible for providing personalized learning to the learners by considering several related learners information such as individual characteristic, learning style, and knowledge background. With context aware computing perspective, this paper thus proposes the context-aware recommendation system to promote effective personalized online learning for each learner individually. Instead of employing ordinary individual context, this paper focuses also on the social context which is the interaction between learning objects and the learners. The gathered social context is classified with K-nearest neighbor and decision tree for classifying appropriate types of learners. Consequently, the appropriate learning paths are recommended by using association rule. The empirical study is conducted with the learners having scientific and non-scientific backgrounds studying in two different content modules of basic computer skill course. The results show that the proposed social context-aware recommendation system is able to provide acceptable classification accuracies from both classifiers. Additionally, the proposed system is potentially able to recommend appropriate learning path to different group of learners.

Application of Topological Network Measures in Manufacturing Systems

Manufacturing systems are complex networks of material flow. Complex network theory has been used as a descriptive and empirical research tool for various network types. However, due to the distinct origin of the various investigated networks (e.g., social networks, biological networks, the Internet), it is not clear if there is a meaningful application of network measures in manufacturing systems. This chapter investigates network modeling and network measures in manufacturing systems, and categorizes them according to their type of research.