This paper presents a quantitative assessment model for the assessment of environmental load in gold production, which is based on environmental ecology, Life Cycle Assessment (LCA) and artificial neural network. In LCA evaluation indexes include resource consumption, energy consumption, gas waste, liquid waste, solid waste and so on. Each item represents an aspect in gold production while their combination reflects the relationship between gold production and environment. In this paper an utility function of evolution index on environmental load is given by linear or non-linear transformation, which changes evaluation indexes with different dimension into an equivalent value in range of [-1, 1]. This paper also provides a new approach to make decision on the weight for quantitative assessment with the help of a three-layered artificial neural network. The proposed approach in this paper has been successfully used to a real synthetic assessment for the environmental load in gold production, which include seven kinds of different operational technologies in China. The applications prove that the new approach is successful and reliable. It provides a new tool for the assessment of environmental load of the gold production.