Development of radiation dose prediction program based on measurement data

Dose assessment is essential for optimizing the radiation dose of workers in nuclear power plants (NPPs). However, in the event of an accident at an NPP, predicting worker doses can be challenging due to limitations such as the unavailability of source term information. To address this, we developed an assessment tool that predicts worker doses using relatively easy-to-obtain dose rate measurements. The objective of this study is to develop a dose prediction program based on measurement data. First, a dose prediction model was designed using the Kriging methodology. Second, the model and user interface were implemented to develop the program using C# and Windows Forms. Finally, the prediction results of the developed program were validated against MCNP code simulations using hypothetical verification scenarios. The prediction model employs the Kriging methodology to analyze spatial correlations among measured data and derive optimal interpolation weights. The program interface comprises three primary modules: (1) data input module, (2) result output module, and (3) work scenario module. It allows users to input measurement data, spatial geometry, and worker paths, and is designed to generate 3D dose maps and comparative dose charts. The validation results against MCNP showed mean absolute percent errors (MAPE) of 18.95%, 16.14%, and 14.70% for measurement densities of 10%, 15%, and 20%, respectively. The results of this study can be utilized to establish optimal work plans and ensure worker safety during NPP accidents.