Sandalwood is an economically valuable tree crop grown for its fragrant heartwood used in pharmaceuticals, cosmetics and religious ceremonies. Optimizing the yield from sandalwood plantations is vital but challenged by the dependence on various interconnected environmental factors like temperature, soil moisture and nutrition. This paper surveys past research on smart Internet of Things (IoT)-enabled systems to monitor sandalwood crop health and machine learning models for predictive yield analytics. The real-time data insights from sensors tracking soil humidity, ambient temperature, leaf-wetness and other parameters correlated with tree growth empower data-driven interventions. Advanced machine learning algorithms, especially deep neural networks, show potential in effectively modelling the complex multivariate interactions between climatic conditions, soil properties and sandalwood yields. However, challenges remain in building representative labelled sandalwood datasets from sensors, maintaining reliable field sensor networks, and accurate multivariate time-series forecasting of yields. Further research to create robust IoT architectures coupled with deep learning predictive models is imperative for realizing precision sandalwood agriculture to improve farmer incomes and industry growth.