<p>Welding is widely used to make assembly of body-in-whites, and it will bring serious environmental challenges to society. The environmental impact of welding is usually evaluated based on the welding seams. Different process parameters (i.e., current, voltage, time, arc height) are adopted for finishing the seams, and it results in different environmental impacts. How to select parameters is a key issue for reducing its environmental impact. The study provides a proposal for an assessment method that considers the consumption of filler materials to evaluate the environmental impacts of the welding process, i.e., gas metal arc welding (GMAW), submerged arc welding (SAW), laser welding (LW), and plasma arc welding (PAW). Then, the potential hotspots identified are filler material consumption, electricity consumption, and shielding gas consumption. To minimize the environmental impact of the welding process, the neural network is used for optimizing the abovementioned process parameters in welding. Results show that the Pareto solution set of welding current, energy consumption, and welding material is obtained, which can minimize the corresponding energy consumption and material consumption. Moreover, the applicability of the assessment method is verified by evaluating the carbon footprint of welding a body-in-white, and the 1.5% carbon footprint can be reduced.</p>