Forest fires constitute a widespread and impactful natural disaster, annually ravaging millions of hectares of forests and posing a severe threat to human life and property. Accurate quantitative prediction of forest fire spread is essential for devising swift risk management strategies and implementing effective firefighting approaches. In response to this imperative, this study introduces a Forest Fire Spread Behavior Prediction (FFSBP) model, encompassing two integral components: the Forest Fire Spread Process Prediction (FFSPP) model and the Forest Fire Spread Results Prediction (FFSRP) model. The FFSPP model involves the prediction of the direction and speed of forest fire spread, achieved through a fusion of the Cellular Automata model and the Wang Zhengfei model. On the other hand, the FFSRP model focuses on forecasting the extent of the burned area, leveraging machine learning methods. To validate the efficacy of the proposed models, a real case study is undertaken using the “3.29 Forest Fire” incident in China. The FFSPP model is validated against this case, while the FFSRP model is assessed using a real fire dataset obtained from Montesinho National Forest Park in Portugal. Results from the validation process reveal that during the natural development period of the “3.29 Forest Fire,” the FFSPP model predicts a burned area of 286.81 hm ² , with an associated relative error of 28.94%. This relative error is notably smaller than those observed in the Farsite and Prometheus fire behavior simulation models. Additionally, the FFSRP model demonstrates commendable predictive performance, particularly in the context of small and medium-sized fire scenarios. These findings underscore the potential of the FFSBP model as a valuable tool in enhancing forest fire prediction accuracy and facilitating more robust risk mitigation and firefighting strategies.

Reverse logistics is convincingly one of the most efficient solutions to reduce environmental pollution and waste of resources by capturing and recovering the values of the used products. Many studies have been developed for decision-making at tactical, practical, and operational levels of the reverse supply chain. However, many enterprises face a challenge that is how to design the reverse logistics networks into their existing forward logistics networks to account for both economic and environmental sustainability. In this case, it is necessary to design a novel reverse logistics network by reconstructing the facilities based on the existing forward logistics network. Multi-level investments are considered for facility reconstruction because more investment and more advanced remanufacturing technologies need to be applied to reduce the carbon emissions and improve facility capacities. Besides, uncertain elements include the demand for new products and return quantity of used products, making this problem challenging. To handle those uncertain elements, a bi-objective stochastic integer nonlinear programming model is proposed to facilitate this novel reverse logistics network design problem with economic and environmental objectives, where tactical decisions of facility locations, investment level choices, item flows, and vehicle assignments are involved. To show the applicability and computational efficiency of the proposed model, several numerical experiments with sensitivity analysis are provided. Finally, the trade-off between the profit and carbon emissions is presented and the sensitive analysis of changing several key input parameters is also discussed.