International Journal of Simulation and Process Modelling (6 papers in press)

Regular Issues

• Simulation-driven deep learning framework for early poultry disease detection using faecal image classification with optimised CNN architectures  
  by Nonita Sharma, Monika Mangla, Manik Rakhra, Baljinder Kaur, Raj Kumar Mohanta, Bijay Kumar Paikaray 
  Abstract: The present investigation aims to devise an optimized Convolutional Neural Network (CNN) framework to identify prominent poultry diseases based on faecal images. The proposed research model uses a tri-convolutional layer architecture to enhance the accuracy of poultry disease classification and improve feature extraction. Three major diseases, namely coccidiosis, salmonella, and Newcastle, have been considered. The prime objective of current research is to achieve early detection by employing advanced deep-learning techniques. The proposed model uses fecal images to identify pathological conditions using the TriConvLayer architecture accurately. In this work, custom CNN models, viz. SoloConvLayer, TriConvLayer, and FiveConvLayer models are used to achieve an accuracy of 97%, 98%, and 98%, respectively. The achieved result advocates the efficacy of the proposed approach. It thus has the potential to revolutionize early disease detection in poultry farming, a major step towards improving animal health and farm productivity.
  Keywords: deep learning; poultry disease detection; convolutional neural networks; CNN; faecal image analysis; poultry health monitoring; disease classification; simulation framework.
  DOI: 10.1504/IJSPM.2025.10073525
   
  
• Synchronising inbound-outbound operations in e-commerce cross-docks: a simulation and experimental analysis  
  by T. Praveen Kumar 
  Abstract: The cross-dock operations of e-commerce are usually characterised by congestion, lengthy turnaround times, and unreliable departure times due to limited staging capacities and poor alignments in the decision of truck turnaround time. This paper constructs a field-calibrated discrete-event simulation implementation model of a cross-dock based in Dubai; it uses the Arena to study how the interaction of the scheduling rules and the staging capacity affects the operational performance. It deploys a 3
  Keywords: discrete-event simulation; factorial experimental design; cross-dock operations; scheduling rules; staging capacity; congestion management; e-commerce logistics.
  DOI: 10.1504/IJSPM.2026.10079693
   
  
• A simulation-based modelling framework for DC voltage control and process level stability in bidirectional AC/DC converters for V2G systems  
  by Ying Zhang, Rui Fu, Linquan Tang, Gaosheng Rong 
  Abstract: Maintaining DC voltage stability is a key challenge in simulation-based modelling of bidirectional AC/DC converters for Vehicle to Grid (V2G) systems. This study develops an integrated simulation and process modelling framework that analytically investigates instability mechanisms and supports model driven control design. Using a small signal modelling process, the proposed framework identifies the dynamic coupling among grid voltage, grid current, and load current, and embeds these disturbance dynamics into a multi-disturbance compensation control strategy. The entire development and validation are conducted within a MATLAB/Simulink simulation environment, forming a reproducible workflow for system-level stability evaluation. Simulation results demonstrate that the framework not only enhances DC voltage stability under varying grid and load conditions but also provides a generalizable process-oriented simulation approach for optimising energy conversion systems.
  Keywords: vehicle to grid; V2G; systems; bidirectional AC/DC converter modelling; DC voltage stability; simulation and verification (MATLAB/Simulink); process modelling and optimisation; small signal analysis.
  DOI: 10.1504/IJSPM.2026.10080470
   
  
• A simulation-oriented framework for power equipment lifespan prediction using glyph-enhanced NER from unstructured maintenance work orders  
  by Zhiwei Li, Zhaoyang Zhang, Wei Wang, Yupu Jiang, Ziming Wei, Shaocheng Qu 
  Abstract: Equipment lifespan prediction is essential for maintenance planning and asset management, but PMS work orders and ERP equipment records are often difficult to integrate because of their unstructured and heterogeneous forms. This paper proposes a data-driven framework that combines process-oriented dataset construction, glyph-enhanced NER, ERP-PMS feature matching, and FT-Transformer-based equipment lifespan prediction. An EPE-MR dataset with 12,059 annotated maintenance work-order samples is constructed, and a glyph-enhanced NER model is developed to extract maintenance entities using contextual, radical, and glyph features. The extracted entities are matched with ERP records to form structured life-cycle observations for equipment lifespan prediction. Experiments show that the proposed NER model achieves an F?score of 0.897 and outperforms conventional, pre-trained, span-based, and large language model baselines. Results on transformers, circuit breakers, and disconnectors demonstrate that the constructed observations provide useful process-level information for equipment lifespan prediction and maintenance decision-making.
  Keywords: power equipment lifespan prediction; unstructured maintenance work orders; named entity recognition; NER; glyph and radical fusion; ERP-PMS feature matching.
  DOI: 10.1504/IJSPM.2026.10081111
   
  
• Crowd behaviour simulation and adaptive path planning based on multi-factor fusion and cross-modal data  
  by Xing Zhou, Huan Fang 
  Abstract: High-density crowds in large public venues (e.g., scenic areas, exhibition halls, amusement parks) critically impair passage efficiency and pose severe safety hazards ranging from congestion to stampede risks. Conventional navigation methods primarily optimize for the shortest distance or maximum interest, neglecting dynamic crowd behaviors and safety-critical constraints. This paper proposes a unified framework integrating multi-factor fusion with cross-modal feature alignment for crowd behavior simulation and safety-aware path planning. The framework encodes spatial structure, crowd density, individual preferences, and crowd dynamics into a coherent representation, and incorporates an adaptive dual-mode path planning algorithm for predicting crowd evolution and generating safety-oriented routes. Experiments in a science and technology museum navigation scenario demonstrate that the proposed method achieves significant improvements over traditional algorithms across interest coverage, congestion mitigation, and path utilization efficiency. These findings validate the framework's efficacy and scalability for multi-scenario crowd behavior simulation and intelligent navigation applications.
  Keywords: agent-based simulation; crowd behaviour modelling; adaptive path planning; cross-modal data fusion; multi-factor decision fusion; safety-aware navigation.
  DOI: 10.1504/IJSPM.2026.10081112
   
  
• Simulation for continuous casting mould level fluctuations prediction via SSA-XGBoost model  
  by Jinxiang Chen, Jiahong Xu, Qiang Zhang 
  Abstract: Continuous casting process simulation often faces the problems of small sample size and high data noise, which results in the poor generalisation of embedded anomaly detection models. A hybrid simulation-machine learning (ML) method is presented to predict continuous casting mould (CCM) level fluctuations. A genetic algorithm-symbolic regression (GA-SR) feature enhancement approach is presented, in which a fitness objective function based on Fisher discriminant ratio (FDR) is designed to generates interpretable features with high separability. A sparrow search algorithm (SSA)-XGBoost model is constructed to predict the CCM level fluctuations, where SSA can adaptively adjust the hyperparameters of XGBoost to maximise detection accuracy in a noisy environment. A continuous casting process dataset with 1,395 samples and 33 features is constructed. The simulation results show that the prediction accuracy of the presented method is about 0.8894, which is better than SSA-XGBoost, XGBoost, and NGBoost.
  Keywords: industrial process simulation; anomaly detection; liquid level in continuous casting mould; CCM; XGBoost; sparrow search algorithm; SSA; symbolic regression.
  DOI: 10.1504/IJSPM.2026.10081460