Natural fibre reinforced composites have shown several advantages over synthetic fibre composites, such as higher strength-to-weight ratio, low cost, low density, low energy consumption in fabrication, and low carbon footprint. However, the high cellulose content in natural fibres and inherent flammability of polymers result in the composites with a high propensity to sustain and spread fire. To meet the fire resistance requirements for the composites’ applications in automotive, acoustic, building and aerospace industries, traditional intumescent flame retardant (FR) additives, such as ammonium polyphosphate (APP) and melamine phosphate, are blended into the polymers, which is an effective solution to reduce flammability. An intumescent FR system mainly consists of three components – an acid source, a blowing agent, and a carbonising agent, which forms a carbonaceous char layer on the sample surface during pyrolysis. Chemically, the charring layer contributes to retaining carbon, reducing the heat of combustion per unit mass of the material. Physically, it forms a barrier that restricts the heat and mass transfer within the condensed phase, protecting the underlying material from incident heat flux and constraining the transport of gaseous pyrolysates, which sustain the flame. The burning behavior of these composites with fire retardant additives involves complex condensed phase and gas phase reactions, resulting in the formation of transient char layer and therefore is difficult to model. In this work, we have developed a numerical model based on National Institute of Standards and Technology (NIST) Fire Dynamics Simulator (FDS) for flax fibre reinforced polypropylene (PP) composite with APP, which can predict the material’s heat release rate (HRR). By systematically performing a series of thermal characterisations, such as thermogravimetric analysis (TGA) and temperature-modulated differential scanning calorimetry (TMDSC), thermo-physical properties of the composite and its constituents were determined for the numerical model. Three-dimensional heat transfer model in FDS was employed to solve for heat conduction within the cone calorimeter setup. The reaction order in the Arrhenius function to govern the composite's pyrolysis reaction rate was calculated using the Coats-Redfern method. The composite samples were manufactured using melt mixing and compression molding. Finally, the numerical model was validated with the experimental results using cone calorimeter, which is a bench-scale tool to measure material flammability, based on heat release rate. Overall this study offers a promising route to predict the burning behaviour of natural fibre-based composites with intumescent flame retardant additives.