<p>Nuclear power plant workers can be exposed to radiation under various working conditions, such as during normal operation, maintenance period, and emergency. To establish an optimal work plan, radiation dose prediction and assessment must be conducted in advance. Radioactivity data are essential to assess radiation dose. However, in some situations, these data may not be available, and analyzing precise activity of radiation source in all nuclear power plant work environments may require significant time and financial resource. Therefore, an algorithm is needed that can swiftly estimate the radioactivity of the radiation sources utilizing the radiation dose rate which are readily available data. The objective of this study is to develop an algorithm for estimating the radioactivity of arbitrary geometry radiation sources within the nuclear power plant workspaces. To achieve this, an algorithm was initially developed to estimate the radioactivity of point sources and arbitrary geometry radiation sources. Subsequently, scenarios for the point sources and cylindrical volume sources were established, and the effectiveness of the algorithm was validated with applying these scenarios. In the radioactivity estimation algorithm for the point sources, the radioactivity was estimated by reverse-calculating the external exposure dose assessment equation from the radiation dose rate data. However, various uncertainties may exist in the radiation dose rate data. Because limitations exist in deriving an exact activity value, a numerical analysis method was employed to obtain an optimal solution. The estimation procedure was divided into five steps: (1) obtaining input parameters, (2) establishing the simultaneous equation, (3) deriving an initial solution set, (4) deriving the approximate solution set, and (5) calculating uncertainty and determining the optimal solution. Additionally, a radioactivity estimation algorithm for arbitrary geometry radiation sources was developed integrating the point source radioactivity estimation algorithm and the point-kernel method. The estimation procedure was divided into two steps: (1) point segmentation of the arbitrary geometry radiation source and (2) estimation of radioactivity for the segmented point source. In the validation results of each algorithm, the error ratio between the actual and the estimated values ranged from approximately 1.19% to 11.2%. The results of this study are expected to be utilized in future for optimizing work plans in nuclear power plant workspaces where radioactivity information is unavailable.</p>