Högskolan i Skövde

With the increase in complexity of industrial systems it becomes more and more challenging to make well-grounded decisions for system design and operation. Following the concept of Virtual Factories with Knowledge-Driven Optimization (VF-KDO), this paper proposes a graph database approach to support knowledge-driven and simulation-based optimization. With the mapping of a VF-KDO ontology to a graph database, competency questions that facilitate traceability, transparency, and group decision making can be answered. This is exemplified with an industrial use case and a scenario form academic education.

Understanding delay conditions and making accurate predictions are essential for optimizing turnaround and taxi times, which in turn reduces fuel consumption and lowers CO2 emissions in airport operations. However, while existing research has explored the impact of various prediction models on airport operations, it often overlooks the performance of Collaborative Decision Making (CDM) variables when discussing delay conditions. The implementation of CDM at major European airports has led to a milestone-based approach within airport operations, particularly in the turnaround operations, segmenting these operations with unique features. The purpose of this paper is to systematically investigate the efficacy of various machine learning techniques, such as linear regression, regression trees, random forests, elastic nets, and multi-layer perceptrons (MLP), in accurately predicting delay categories within the CDM framework. For this purpose, we analyzed CDM operational data from Madrid Airport, with at least 166,185 flight observations. Our findings illustrate a training methodology on how different models vary in prediction accuracy when applied to CDM operational data. We applied the SHAP (SHapley Additive exPlanations) method for feature importance analysis of all our independent variables to interpret the output of our machine learning models. Our results indicate that linear regression and elastic nets are the most effective machine learning models for achieving high prediction accuracy within the CDM framework. To test their robustness, we extended the analysis with predictions for better schedule times for taxi times on arrival and depature for selected runways using a different dataset. Our results contribute by showcasing a training methodology, highlighting how elastic net model as the best-performing model can be adopted for turnaround operations. In conclusion, we discuss the implications of our results for runway demand policies and use of airport resources such as gate & runaway allocation. 

Virtual manufacturing, simulation, and optimization provide a wealth of knowledge about the possibilities of future production systems so as to support decision makers. However, this knowledge usually remains with a handful of domain experts, is not captured and is hard to share even within the same team. At the same time, simulations can benefit from the incorporation of linked data from real factories once a process is running. Graph databases provide a possible approach to storing and managing this form of interrelated heterogeneous data, with powerful querying capabilities that can identify important or interesting patterns that might otherwise remain hidden. Current research focuses on one or two aspects of this problem but does not address all at once, despite the potential benefits of the combination. This paper provides a broad literature review of the current directions within research with a special focus on how graphs can support finding knowledge within Virtual Factories, used by larger teams for industrial planning and optimization.

Batching and scheduling orders for ovens in manufacturing is a typical combinatorial optimization problem, and it is critical for production efficiency and customer satisfaction. Customized settings and randomness happened in practical production including the maximum waiting time of an order, capacity thresholds, and availabilities of ovens make the problem more complex. In this paper, we build a simulation model in ten ovens for a heat-treatment process line of cutting tools according to real data from an industrial case, and the model embeds a detailed control logic that integrates existed scheduling methods with various dispatching rules and parameters. After determining an optimal number of running ovens in the production line by simulation-based optimization, we then propose a multi-objective optimization (MOO) enhanced deep reinforcement learning (DRL) approach to schedule orders for the ovens. The DRL agent learns to use an appropriate dispatching rule at a scheduling time point to select orders and formulate a batch. Along with reducing the average tardiness of the orders and maximizing the average effective utilization of the ovens, we find that the explored MOO enhanced DRL approach is more flexible and effective than the heuristic method. 

Ergonomics evaluation methods are typically used for assessing risks of work-related musculoskeletal disorders (WMSDs) and the physical well-being of workers. Traditionally, these methods rely on assessors observing workers performing tasks and assessing potential risks based on observational ergonomics evaluation methods like the Rapid Entire Body Assessment (REBA) and the Rapid Upper Limb Assessment (RULA). While observational methods provide a structured risk assessment framework, they often depend on subjective evaluations, leading to inconsistent assessments between different ergonomists.

This study examines the application of motion capture technology to enhance the objectivity and efficiency of ergonomics evaluations and to enable the use of direct measurement ergonomics evaluation methods. The study was conducted at a medium-sized enterprise assembly station, where a worker’s tasks were recorded using motion capture technology. The captured motions were input into the Digital Human Modelling (DHM) tool IPS IMMA, and ergonomic assessments were performed using RULA, REBA, and the Arm Force Field (AFF) method.

The study followed a process comprising three main stages: data collection, data processing, and ergonomics evaluation. The recorded data were processed into XML format, imported into IPS IMMA, and exported to the Ergonomics in Production Platform (EPP) for RULA and REBA evaluations and to a script for AFF evaluations. The integration of these methods improved the precision and reliability of ergonomics assessments by replacing subjective estimates with direct measurements. The findings demonstrate the potential of combining motion capture with DHM tools to enhance ergonomics evaluation and support decisionmaking in workstation design and automation.

Owing to the realization of advanced manufacturing systems, manufacturers have more flexibility in improving their processes through design decisions. Design decisions in production lines primarily involve two complex problems: buffer and resource allocation (B&RA). The main aim of B&RA is to determine the best location and size of buffers in the production line and optimally allocate production resources, such as operators and machines, to workstations. Inspired by a real-world case from the marine engine production industry, this study addresses B&RA in high-mix, low-volume hybrid flow shops (HFSs) with feed-forward quality inspection. These HFSs can be characterized by uncertainties in demand, material handling, processing times, and quality control. In this study, the production environment is modeled via discrete-event simulation, which reflects the features of the actual system without requiring unreasonable or restrictive assumptions. To replace the expensive simulation runs, five widely used regressor machine learning algorithms in manufacturing are trained on data sampled from the simulation model, and the best-performing algorithm is selected as the predictive model. To obtain high-quality solutions, the predictive model is coupled with an enhanced non-dominated sorting genetic algorithm (En-NSGA-II) that incorporates lifelong meta-learning and features a customized representation and a variable neighborhood search. Additionally, a post-optimality analysis using a pattern-mining algorithm is performed to generate knowledge for allocating buffers and operators based on the optimization results, thus providing promising managerial insights.

In today’s advanced manufacturing landscape, optimizing production processes is crucial for maintaining competitiveness. Among various optimization challenges, production sequencing in make-to-order hybrid flow shops (HFSs) stands out as particularly complex. This study investigates production sequencing in an HFS from the marine engine production industry, characterized by feed-forward quality inspection (FFQI). In FFQI, rejected engines must be repaired rather than scrapped. The complexity is further heightened by the fact that repair capacity is usually limited to a few engines and rejection at quality inspection leads to sequence scrambling at downstream stations. To address this issue, this study employs simulation-based, knowledge-driven optimization that utilizes real-world data on the rejection rates of different engine variants. This data is used to cluster the variants into three groups with different risks of rejection at quality inspection, informing production sequencing decisions. A non-dominated sorting genetic algorithm, enhanced with anti-block (AB) and anti-delay (AD) strategies (NSGAIIAB-AD), is developed to optimize throughput and delivery delay. AB aims to mitigate the succession of high-risk product variants, minimizing blockage probabilities in the quality inspection stage. AD prioritizes engines with earlier due dates from the same risk category to prevent unnecessary delivery delays. The study also evaluates the impact of extending planning horizons beyond the current 3-day standard. Results demonstrate the effectiveness of the AB and AD strategies, yielding a 10% improvement in average current throughput. Moreover, adopting a 5-day planning horizon leads to an 18% decrease in average delay compared to the current 3-day horizon.

Harvest scheduling and transport are crucial for the delivery performance of a wood supply chain, ensuring that product volumes are delivered on time and in the right quality. This paper suggests three delivery performance objectives for the wood supply chain: service level, lead time, and throughput. It presents a framework for optimizing these objectives by finding trade-off solutions using simulation-based multi-objective optimization. Due to the complexity of the wood supply chain, discrete-event simulation is used to evaluate delivery performance from harvesting to customer delivery. The harvest scheduling problem is formulated as a permutation optimization solved by a customized NSGA-II algorithm with a comparison of three crossover mechanisms implemented: Random Key Simulated Binary Crossover, Order Crossover, and Partially Mapped Crossover, specifically designed for general forestry permutation optimization problems. Analyzed with a heatmap for the visualization of the mapping of the decision space to the Pareto-optimal solutions, the results indicate that the Partially Mapped Crossover performs best. Other simulation-optimization generated data are processed and visualized in an interactive, web-based dashboard for decision-makers, such as forest managers, allowing them to analyze meta-heuristically optimized solutions in both the solution and decision spaces, guiding them to find the most suitable harvest schedules. 

The formulation and selection of parameters and sequences in crankshaft production present challenges that are both demanding and time-intensive. This study introduces an innovative approach to intelligent process planning in crankshaft machining lines using multi-turret machines. Emphasis is placed on automating process planning through multi-objective optimization of critical decisions such as process parameters, operation sequencing, and tool positioning on turret magazines. The principal objectives addressed include minimizing machining and non-machining time, reducing costs by optimizing tool life, and enhancing product quality through optimal surface roughness. By automating these decision points, the proposed framework reduces manual intervention and aligns with Industry 4.0 goals for adaptive, data-driven manufacturing. Additionally, we discuss the potential future incorporation of artificial intelligence agents to dynamically refine parameters and enable adaptive planning.

In recent years graph databases (GDBs) have become popular for managing large amounts of data. Especially the capability to integrate and link heterogeneous data is a driving factor for its success. This makes it a promising tool to support knowledge driven optimization (KDO), which involves various types of data, information, and artifacts. In particular, the link between simulation model, optimization algorithms with their parameters, results, and preferences of the stakeholders is interesting for investigation and traceability. In this paper, we propose a simple ontology as a schema for a KDO graph database and demonstrate how it can be used with an industrial use case. Furthermore, we highlight the ability to add lower level model information from an industrial use case to the GDB.