Artificial Intelligence (AI) has permeated a variety of fields, revolutionising task automation with its advanced algorithmic solutions. However, a notable challenge persists in its struggle with complex computational tasks, often leading to oversimplified and imprecise outcomes. Addressing these limitations, Multi-Strategy Learning (MSL) emerges as a comprehensive approach, integrating multiple inferential and representational strategies to enrich the learning process. This paper presents a novel application of MSL in the context of bicycle imbalance forecasting within a university campus environment. By harnessing the power of MSL, the study aims to enhance the predictive accuracy of AI systems while incorporating the principles of explainable AI (XAI) to demystify AI decision-making. The proposed approach is designed to interpret complex data inputs and provide transparent, actionable insights for campus transportation management. This research not only demonstrates the effectiveness of MSL in addressing real-world problems but also sets a new benchmark for AI applications that require both high accuracy and explainability. The ensemble model performance was: MSE 17.0457, MAE 2.6263, RMSE 4.1286, R-2 0.6751 and XAI feature importance analysis showed Wind Speed, Dry bulb, Wet bulb, and Hour 16 as the top features affecting the model prediction.