Background: Systematic literature reviews (SLRs) build the foundation for evidence synthesis, but they are exceptionally demanding in terms of time and resources. While recent advances in artificial intelligence (AI), particularly large language models, offer the potential to accelerate this process, their use introduces challenges to transparency and reproducibility. Reporting guidelines such as the PRISMA-AI (Preferred Reporting Items for Systematic Reviews and Meta-Analyses–Artificial Intelligence Extension) primarily focus on AI as a subject of research, not as a tool in the review process itself. Objective: To address the gap in reporting standards, this study aimed to develop and propose a discipline-agnostic checklist extension to the PRISMA 2020 statement. The goal was to ensure transparent reporting when AI is used as a methodological tool in evidence synthesis, fostering trust in the next generation of SLRs. Methods: The proposed checklist, named PRISMA-trAIce (PRISMA–Transparent Reporting of Artificial Intelligence in Comprehensive Evidence Synthesis), was developed through a systematic process. We conducted a literature search to identify established, consensus-based AI reporting guidelines (eg, CONSORT-AI [Consolidated Standards of Reporting Trials–Artificial Intelligence] and TRIPOD-AI [Transparent Reporting of a Multivariable Prediction Model of Individual Prognosis or Diagnosis–Artificial Intelligence]). Relevant items from these frameworks were extracted, analyzed, and thematically synthesized to form a modular checklist that integrated with the PRISMA 2020 structure. Results: The primary result of this work is the PRISMA-trAIce checklist, a comprehensive set of reporting items designed to document the use of AI in SLRs. The checklist covers the entire structure of an SLR, from title and abstract to methods and discussion, and includes specific items for identifying AI tools, describing human-AI interaction, reporting performance evaluation, and discussing limitations. Conclusions: PRISMA-trAIce establishes an important framework to improve the transparency and methodological integrity of AI-assisted systematic reviews, enhancing the trust required for the responsible application of AI-assisted systematic reviews in evidence synthesis. We present this work as a foundational proposal, explicitly inviting the scientific community to join an open science process of consensus building. Through this collaborative refinement, we aim to evolve PRISMA-trAIce into a formally endorsed guideline, thereby ensuring the collective validation and scientific rigor of future AI-driven research.

Machine vision solutions can perform within a wide range of applications and are commonly used to verify the operation of production systems. They offer the potential to automatically record assembly states and derive information, but simultaneously require a high effort of planning, configuration and implementation. This generally leads to an iterative, expert based implementation with long process times and sets major barriers for many companies. Furthermore the implementation is task specific and needs to be repeated with every variation of product, environment or process. Therefore a novel concept of a simulation-based process chain for both—configuration and enablement—of machine vision systems is presented in this paper. It combines related work of sensor planning algorithms with new methods of training data generation and detailed task specific analysis for assembly applications.

Detecting anomalies in image data plays a key role in automated industrial quality control. For this purpose, machine learning methods have proven useful for image processing tasks. However, supervised machine learning methods are highly dependent on the data with which they have been trained. In industrial environments data of defective samples are rare. In addition, the available data are often biased towards specific types, shapes, sizes, and locations of defects. On the contrary, one-class classification (OCC) methods can solely be trained with normal data which are usually easy to obtain in large quantities. In this work we evaluate the applicability of advanced OCC methods for an industrial inspection task. Convolutional Autoencoders and Generative Adversarial Networks are applied and compared with Convolutional Neural Networks. As an industrial use case we investigate the endoscopic inspection of cast iron parts. For the use case a dataset was created. Results show that both GAN and autoencoder-based OCC methods are suitable for detecting defective images in our industrial use case and perform on par with supervised learning methods when few data are available.

Aircraft engines are regularly inspected with borescopes to detect faults at an early stage and maintain airworthiness. A critical part of this inspection process is accurately measuring any detected damage to determine whether it exceeds allowable limits. Current state-of-the-art borescope measurement techniques—primarily stereo camera systems and pattern projection—face significant challenges when engines lack sufficient surface features or when illumination is inadequate for reliable stereo matching. MEMS-based 3D scanners address these issues by focusing laser light onto a small spot, reducing dependency on surface texture and improving illumination. However, miniaturized MEMS-based scanner borescopes that can pass through standard engine inspection ports are not yet available. This work examines the essential steps to downsize MEMS 3D scanners for direct integration into borescope inspections, thereby enhancing the accuracy and reliability of aircraft engine fault detection.

Deploying deep learning-based applications in specialized domains like the aircraft production industry typically suffers from the training data availability problem. Only a few datasets represent non-everyday objects, situations, and tasks. Recent advantages in research around Vision Foundation Models (VFM) opened a new area of tasks and models with high generalization capabilities in non-semantic and semantic predictions. As recently demonstrated by the Segment Anything Project, exploiting VFM's zero-shot capabilities is a promising direction in tackling the boundaries spanned by data, context, and sensor variety. Although, investigating its application within specific domains is subject to ongoing research. This paper contributes here by surveying applications of the SAM in aircraft production-specific use cases. We include manufacturing, intralogistics, as well as maintenance, repair, and overhaul processes, also representing a variety of other neighboring industrial domains. Besides presenting the various use cases, we further discuss the injection of domain knowledge.

Visual surface defect detection in complex industrial inspection scenarios remains predominantly manual. Deep learning–based vision systems promise to automate this task but require large quantities of representative, annotated defect images, which are often unavailable due to the rarity and diversity of relevant defects. We address this challenge with a modular synthetic data generation pipeline leveraging physically based 3D rendering, which enables systematic domain randomization throughout the entire rendering chain. The pipeline comprises dedicated modules for scene composition, geometry-based defect generation, procedural material modeling, rendering, and automatic pixel-wise annotation, allowing users to expose and tune key parameters such as defect size, position, and penetration depth, object materials, camera poses, and lighting. We first provide a structured analysis of all process steps that influence image appearance in defect synthesis and then instantiate the pipeline for a case study on aircraft engine borescope inspection, targeting missing-material defects on high-pressure compressor blades. Using a two-stage experimental design, we employ a Plackett–Burman screening over 22 candidate factors followed by a one-factor-at-a-time optimization of the most influential parameters. Our results show that the material model, defect positioning, defect size, and penetration depth randomization dominate detection performance. Training an object detector solely on synthetic images generated with the optimized configuration yields a maximum of 0.89 mAP@0.5IoU on a real-world borescope validation set, with performance saturating at ∼500 rendered training images. These findings demonstrate that carefully structured domain randomization in a physically based rendering pipeline can provide practically useful synthetic training data for industrial defect detection.