Aflatoxin B1 (AFB1) is a secondary metabolite produced by fungus of genus Aspergillus. It has been widely found in food, crops and animal feeds which are not properly stored. Aflatoxin B1 is highly carcinogenic even if its concentration is at ppb levels. AFB1 contamination in food, crops and animal feeds has raised much concern during recent years and many methodologies for AFB1 determination in these commodities have been developed. However, little attention was paid to potable water which is the focus of this project. In fact, potable water can be another potential source of AFB1 contamination because of the detected existence of Aspergillus in the water. The aim of this project was to develop a fast and reliable methodology for AFB1 determination in potable water. Lateral flow immunoassay (LFIA) is a very popular bioassay format that receives various applications in the food contaminants determination, environmental monitoring as well as clinical diagnosis. Compared to conventional detection methodologies such as high-performance liquid chromatography and enzyme-linked immunoassays, it offers an inexpensive and relatively rapid assay format. It can be used by people without professional training, which makes it popular in areas where no sophisticated laboratory equipment is available. As a result, LFIA was selected as the target method for AFB1 determination in this project. To reach this aim, research was carried out to optimize the experimental parameters of LFIA in order to achieve good sensitivity and long shelf life. The research mainly contains following steps: labelling primary antibody with gold nanoparticles, fabricating lateral flow strip, optimization of sensor parameters, measurements of potable water samples, result verification and evaluation of sensor stability. Final results indicated that the amount of antibody during the conjugation process has the most significant influence on assay sensitivity and signal intensity. It was also found that magnesium and calcium ions in potable water interfered with the assay, and addition of 0.1% ethylene diamine tetra-acetic acid (EDTA) showed excellent performance in reducing such matrix effect. The device reached 0.5ppb of visual detection limit for most of the water samples tested and it maintained satisfactory sensing performance for three months when stored in refrigerator.