SUMMARYTo investigate the differences of thermal response between heat flux sensors and human skin on the flame manikin, a three-dimensional heat transfer model was developed and validated by the flame manikin system. The initial temperature of the model with sensor material was set to 300 K, and the model with skin material was set as the real condition. Simulated results validated the effectiveness of heat flux measured by the sensor. The incident heat flux through the measured surface was influenced by the different emissivity of the human skin and experimental sensors. Significant difference was found for the temperature response of these two kinds of materials within 4-s fire exposure. The heat flux measured by sensor or the simulated results with actual human skin parameters could be used as the input boundary condition of the skin heat transfer model for Henriques's skin burn prediction. It is necessary to study the actual skin thermal response by experiments, where the 3D model established in this study could be used as the supplementary means for skin simulant sensor development. These findings will also be adopted in our following study of skin burn prediction module in the 3D full-scale simulation platform. Copyright © 2016 John Wiley & Sons, Ltd. Fire-resistant protective clothing is the only barrier that the firefighters could rely on to decrease thermal hazards [2]. Thermal protective performance of the fireproof fabric or garment is a crucial index, which could represent its protective capacity for firefighters. Bench-scale tests, i.e., TPP test [3], RPP test [4], and full-scale flame manikin test [5,6], are the advanced methods to evaluate the thermal protective performance of the fire-resistant fabrics and garments. Burn injuries or time required to skin burns are commonly predicted by these systems according to the temperature or cumulative amount of energy measured by the thermal sensor.Literatures indicate that many thermal sensors are capable of predicting the heat flux transmitted through human bodies under fire exposures, i.e., copper calorimeter, embedded sensor and skin simulant sensor [7]. The heat flux is calculated generally depending on the temperature rise, thermodynamic properties and construction of the sensor, where the temperature rise is closely related to the sensor properties.As for the bench-scale tests, fire-resistant fabric is usually put in the middle of the heat source and test sensor; hence, the sensor could measure the temperature on the backside of the fabric. The cumulative Fire Mater. 2016 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002DOI: 10. /fam.2363 energy exposure value could be calculated to predict skin burns [8,9]. Figure 1(a) shows the thermal sensor used in the bench-scale test, which is the thermal capacitance calorimeter with copper slug recommended in ASTM E457 [10]. The exposed surface of the copper slug calorimeter is paint with a thin black coating with an absorptivity of 0.9 or greater, which is designed to absorb ...