Overgrown roadside vegetation poses a danger to road users by obstructing the visibility of the road and potentially obscuring signs or other traffic participants. Thus, regulations clearly define the height above the road that must not be obscured. As manually identifying such incidents is time-consuming, we propose 3D Clearance Control: a pipeline that automatically detects vegetation in need of trimming. Our system is based on LiDAR point clouds, which give access to accurate position and height information. It comprises four main steps: the semantic segmentation of the point cloud, the aggregation of scans within a scene, the estimation of road boundaries, and the creation of the volume representing the clearance gauge. We developed a modular process to perform a comprehensive evaluation combining different segmentation models and road boundary approximation methods. We mea-sure the accuracy and computing times on three widely used street-level datasets: SemanticKITTI, NuScenes, and PandaSet. We achieved an mIoU of 67.6 on our annotated test scenes and a speed increase of 52.7% compared to previous systems.

With the increase in data sharing and a growing number of data consumers, access control can become a challenging task to manage. This paper introduces the development of the Finer-Grained Attribute-based Policy (FiGAP) machine, an innovative framework designed to manage access within a Data Trust. FiGAP leverages fine-grained access control mechanisms with inheritance to facilitate enhanced data sharing across different organizational boundaries. Unlike other access control models, this paper develops and thoroughly explains a distinct framework that allows for the establishment of an adaptable Data Trust, potentially as a service. Additionally, this approach also provides an exceptionally detailed level of permission control.

The growth of e-commerce has intensified challenges in the last-mile delivery, which represents a critical and costly segment of the supply chain. The increasing demand for integrative and sustainable supply solutions highlights the need for approaches that balance the environmental, social, and economic dimensions of last-mile delivery. A promising solution involves consolidating deliveries with zero-emission vehicles to central, barrier-free pick-up stations in urban outskirts, reducing environmental impact while addressing accessibility issues. Simulation-based approaches are emphasized as a tool to address the last-mile delivery challenges of urban outskirts. This paper presents a systematic literature review on simulation-based approaches to sustainable and inclusive last-mile delivery in urban outskirts. It addresses the underexplored logistical challenges in these areas and examines how customer needs impact on the three pillars of sustainability: social inclusion, ecological responsibility, and economic viability. By synthesizing current peer-reviewed research from 2020 to 2025, this study identifies critical research gaps, particularly with regard to designing inclusive infrastructures and developing adaptive simulation tools for rural areas and urban outskirts. This work contributes to a comprehensive research agenda, offering insights into inclusive, efficient, and environmentally conscious last-mile delivery solutions for urban outskirts.

The application of developments in the real world is the final aim of all scientific works.In the case of Data Science and Machine Learning, this means there are additional tasks to care about, compared to the rather academic part of "just" building a model based on the available data.In the well accepted Cross Industry Standard for Data Mining (CRISP-DM), one of these tasks is the maintenance of the deployed application.This task can be of extreme importance, since in real-world applications the model performance often decreases over time, usually due to Concept Drift.This directly leads to the need to adapt/update the used Machine Learning model.In this work, available model-agnostic model update methods are evaluated on a real-world industry application, here Virtual Metrology in semiconductor fabrication.The results show that for the real-world use case sliding window techniques performed best.The models used in the experiments were an XGBoost and Neural Network.For the Neural Network, Model-Agnostic Meta-Learning and Learning to learn by Gradient Descent by Gradient Descent were applied as update techniques (among others) and did not show any improvement compared to the baseline of not updating the Neural Network.The implementation of the update techniques was validated on an artificial use case for which they worked well.

The paper aims to create an understanding of the importance of remote sensing data and its efficient analysis. The authors show the benefits that the combination of unmanned aerial vehicles and deep learning algorithms can create. The focus of the paper is on a practical example: using a semi-supervised neural network and remote sensing data to distinguish between healthy and unhealthy trees. This scenario occurs in similar contexts in the real world and can provide added value for the observation of fruit trees and forest stands. Particularly in view of climate change and its effects, it is extremely important to ensure good health monitoring of ecosystems in order to initiate appropriate measures at an early stage. An Open-Source-framework serves as the basis. It is based on a semi-supervised approach and promises to create a good neural network with comparatively little annotated data. This work shows how Deepforest can be extended from simple tree detection to multi-class health monitoring.

This paper presents an investigation into the economic impact of drones in viticulture, an area that has not been previously researched. The authors calculate the economic impacts of drones in viticulture and use this to measure the overall sustainability impact assessment for this technology. The study aims to explore a method for investigating the impact of technological changes on business processes in viticulture. This involves selecting viticultural business processes, representing them using Business Process Model and Notation (BPMN), and considering their simulation. Backpack sprayers and trailed sprayers were considered conventional application methods. The application of crop protection products by drone was considered a digitalized variant. Literature research and guideline-based expert interviews with vinegrowers provided the information basis. The study focuses on plant protection in viticulture, but the results can be applied to other agricultural processes. By integrating all three indicators for sustainability, the study provides an evidence-based method for evaluating the impact of drones on viticultural business processes.

Background: Future predictions of the world population and the demand for agricultural products per capita suggest that we have to increase food production at least two-fold by 2050 despite negative forces such as climate change with more frequent extreme weather conditions, degradation of land, increasing land scarcity, shortness of water, desertification, fooding and the loss of biodiversity for ecosystem services. One pathway how to achieve this challenge is sustainable intensification, which formulates a coarse goal rather than a detailed guideline. A heterogeneous compound of strategies is necessary and a central aspect involves examining the potential of AI technology to boost production efficiency while mitigating negative environmental consequences. Methodology: To comprehensively examine this topic, two systematic literature searches were conducted. The first aimed to gather information on sustainable agriculture and the environmental costs of conventional practices. The second focused on identifying explicit AI applications and their impact. Results: Numerous examples demonstrated how sustainable AI development drives agriculture towards a more sustainable future. The main contribution of this study is the Data-Model-Purpose matrix (DMP matrix) and the derived Bayesian matrix for a comprehensive analysis of several AI applications in agriculture and their relations to data sources and algorithms.

Near-real-time monitoring and classification of business process event streams is becoming more and more prominent. This also includes ensuring data quality for the application of downstream online process mining activities and therefore identify and classify incorrect process behavior of incoming event streams in an online setting, what is considered too little in existing approaches. In this paper, we present an online classification approach that supports monitoring and anomaly detection in event streams at the event level. Possible process drifts can be handled by an online learning workflow. By integrating two explanatory components, the results of the online classification are made transparent and comprehensible. Through a technical experiment, the performance of the classification approach is evaluated based on different data sets. Thereby, the classification model achieves an average F1 score of 0.877 with an average processing time of ∼15 ms per event.

The analysis of business processes has become increasingly important in recent years, not least due to the emergence of analysis tools that enable data-centric views of processes and thus provide increasingly operational support for process flows. In this work, a semi-supervised classification model is presented that takes into account different developments in deep learning (e.g., deep generative models), time series analysis (e.g., long short-term memory) and sequence processing (e.g., attention mechanism) and combines them in one approach. The results of the experimental implementation of the classification model show that it is able to filter activity-related and time-related anomalies from the event data and outperform existing approaches in its classification accuracy (F1 score). The classification model achieves an F1 score of up to 93%.

The production of food in an ecologically and economically sustainable manner is of significant importance today. Agricultural producers are increasingly being accompanied by elements of Agriculture 4.0 such as automation and decision-making support. This work shows an example of how the digitization of viticulture can be significantly supported by Deep Learning. The work presents an approach that can overcome the loss of human expertise in grape identification by using image-recognition-techniques and residual network architectures. Our developed model for grape identification at a vineyard reaches an accuracy of 99% of correctly recognized grape varieties.

Anomaly-based Intrusion Detection Systems (IDS) can be instrumental in detecting attacks on IT systems.For evaluation and training of IDS, data sets containing samples of common security-scenarios are essential.Existing data sets are not sufficient for training modern IDS.This work introduces a new methodology for recording data that is useful in the context of intrusion detection.The approach presented is comprised of a system architecture as well as a novel framework for simulating security-related scenarios.

The irreversible process of demographic change, especially in Germany, leads to numerous challenges.According to this, research has to face the task to integrate the constantly ageing population into the urban and public space in such a way that there are as few barriers as possible.With the support of digitalization, so-called smart urban objects are being designed in order to do make integration, so that people and the available technology can be used most efficiently.A special ontology has been developed to meet this demand.

For decisions on operational business level it is necessary to be aware of and to understand what happens around you and what probably will happen in the near future. E.g. in Operating Room Management, and especially in Online surgery scheduling, a lot of decisions are difficult to handle, since there are high cognitive and communicational efforts to gather all necessary information to oversee the current situation and to derive decisions based on this information. However, the emerging trend of connecting devices and new methods in data analytics, allow new approaches for decision support in these areas. By using this concepts we suggested and implemented a data-driven architecture including several components for data analysis, information generation and visualization. The presented prototype means a proof-of-concept of the idea of a real-time and situation-aware decision support system for operating room managers.

The analysis of surgical activities became a popular field of research in recent years. Various methods had been published to detect surgical phases in various data sources in the operating room. Objective of this research is to develop a method for utilizing real-time information to extract surgical activities. In this work we use fine-grained data of surgical devices and operating room equipment which is produced permanently during surgeries. This low-level data help describing the current surgical phases and reflect real-time status of the endoscope, insufflator, electrosurgical devices and light sources. This is the basis for the development of a structured process to extract surgical phase recognition models. We show how to integrate expert knowledge and transfer this information into an automated and scalable information system for surgical phase recognition. The artifact is developed by adapting the method engineering methodology to find a best practice for utilizing fine-grained data for intrasurgical activity detection. We evaluated our approach with 15 data sets of laparoscopic surgeries and obtained an accuracy rate of about 83% with this approach.

Collaboration in business environments is an ongoing trend that is enabled by and based on cloud computing.It supports flexible and ad-hoc reconfiguration and integration of different services, which are provided and used via the internet, and implemented within business processes.This is an important competitive advantage for the participating stakeholders.However, trust, policy compliance, and data privacy are emerging issues that result from the distributed data handling in cloud-based business processes.Up to now, several architectures and technical systems that enable the cloud-based collaboration within business processes have been developed, but the selection of an appropriate business process management system (BPMS) is missing.An implemented BPMS has to meet certain requirements that result from the cloudbased characteristics and from the other implemented systems.This paper derives requirements for BPMSs in cloud-based environments, currently available BPMSs are evaluated against the derived requirements and the selected one is implemented subsequently.

In the digital age, where research is data-driven, understanding all involved fields of research becomes more and more important. Understanding various data sources within interdisciplinary research and beyond domain boundaries is a significant core competency. All participants should have a same-level understanding of significant information, which can be created from various data sources. Based on this fact, the paper at hand demonstrates a modeling approach for the generation of a unified data model in terms of smart urban objects. These smart objects are represented by interconnected data structures which is a prime example in context of Internet of Things. Further, an implementation of the graph database Neo4J and a correlated visualization of intuitive structuring of data sources beyond domain boundaries will be demonstrated.

One of the contemporary problems, and at the same time a big opportunity, in business networks of supply chains are the issues associated with the vast amounts of data arising there.The data may be utilized by the decision support systems in supply chains; nevertheless, often there are information privacy problems.The supply chains in cloud will need appropriate administration for support of privacy aspects of cooperating business units existing in big data ecosystems.In this paper we analyze the possibility of utilizing the big data technology for supporting business processes optimization with respect of the privacy regulations in supply chains under the usage of the big data analytics lifecycle.We present our approach on an example of a business process in logistics.

Today's business is more and more organized in collaborative networks.Although decision makers know the benefits of collaboration, they are afraid of losing control of their data, which is one of the main impediments for Cloud Computing.We propose a novel cloud based approach for collaboration in business processes with guaranteed control of the privacy of the data.The platform ensures the compliance with the companies' privacy policies and laws.The paper shows the definition of privacy policies and how they are converted into a well established access control language.An example helps to clarify the methods.

In this paper we present an approach for an information system which is capable of processing and analysing vast amounts of data. In addition to Big Data solutions we do not focus on ex post batch processing but on online stream processing. We use Apache Storm in combination with Complex Event Processing to provide a scalable and dynamic event-driven information system, providing logistics businesses with relevant information in real-time to increase their data and process transparency.

The emergence of a closer relationship between cloud service providers in the cloud computing market is the inevitable consequence of the computing as utility concept.The closer cooperation creates competitive advantages for providers and users of cloud services as well.Capacities and services can be used in a collaborative and flexible way.Despite the numerous potentials of composite cloud services, trust, policy and privacy are the major challenges resulting from the distributed and flexible data handling.The paper derives requirements and solutions in the field of inter-cloud service communication with a special focus on security.The proposed architecture is evaluated with a sample collaborative business process of inter-cloud service interaction.

Logistics and its involved parties are nowadays faced with demanding challenges, in order to fulfill their customers' needs.Hence logistics service providers are constrained to cooperate with each other, which leads to the challenge of 'understanding' each other's service descriptions and integrating the strongly differing IT-systems.An emerging approach to solve this problem is the operation of cloud platforms.Main tasks are service retrieval and composition.However, a suitable visualization is needed to generate a high user acceptance.With the service map concept a first step is taken, that certainly needs further improvement.This paper briefly gives an introduction to general information visualization and analyzes the suitability of several approaches for improving the service map concept with regards to different scales of measurement.After their comparison a general guideline for fostering the visualization concept is derived.Objective is the increase of information content while keeping an intuitive usability.

In recent years many efforts were dedicated to the elicitation and definition of requirements for software development projects. However, by concentrating requirement discussions on software development there is a tendency to neglect the predominant role of standard application software (SAS) in enterprises. The process of choosing a SAS product is poorly understood, frequently ill-structured, and performed in an inefficient and ineffective way leading to suboptimal results. Focusing functional aspects, this paper suggests an approach for describing business requirements and software characteristics in terms of ontologies. This way a formal representation of both, requirements and software features, becomes available and can be used for semi-automated reasoning about the suitability of a certain product. The paper contributes directly to the requirements engineering research and addresses a widely ignored topic by suggesting a solution proposal for the above sketched problem.

Based on the investigation of the international state of the art of structures for first and second generation knowledge based systems, distributed artificial intelligence (AI) is considered. For the modularization of second generation knowledge based systems architecture criterions, aspects of distributions and available concepts for the communication of distributed system modules (agents) are presented. In addition a three level CORBA based architecture for distributed AI systems in a heterogeneous network is discussed. The architecture consists of a knowledge client level, a knowledge domain agent server level and a persistent knowledge storage level supported by a semantic/presentation split of logical knowledge objects. The proposed architecture completes the advantages of standardized communication protocols such as CORBA with the productivity of object oriented database management systems. First results of the implementation are presented in the system VISIS using VisualWorks/Smalltalk (ParcPlace) and Distributed Smalltalk (HP).>