To determine the optimal warehouse location, it is usually assumed that the collected data are uncontaminated. However, this assumption can be easily violated due to the uncertain environment and human error in disaster response, which results in the biased estimation of the optimal warehouse location. In this study, we investigate this possibility by examining these estimation effects on the warehouse location determination. Considering different distances, we propose the corresponding estimation methods for remedying the difficulties associated with data contamination to determine the warehouse location. Although data can be contaminated in the event of a disaster, the findings of the study is much broader and applicable to any situation where the outliers exist. Through the simulations and illustrative examples, we show that solving the problem with center of gravity lead to biased solutions even if only one outlier exists in the data. Compared with the center of gravity, the proposed methods are quite efficient and outperform the existing methods when the data contamination is involved.