Machine learning, concerns the construction and study of systems that can learn from data. Learning is the fundamental and most important element of biological intelligent systems. We use methods of intelligent system in industrial process. Moreover, with fractal geometry, we analysed the complexity of robot laser-hardened patterns. We analysed patterns hardened with different robot laser cell parameters; namely, the two parameters of speed and temperature. Fractal dimensions were calculated with a method for estimating the Hurst exponent H for 3D object. Also, fractal dimensions were used for pattern recognition and for describing the roughness of the hardened patterns. However, owing to the complexity of real-world problems, the development of intelligent learning systems still remains a challenging topic. For the analysis of results, we use an intelligent system method, namely, a neural network, genetic programing and regression. The general goal of Artificial Intelligence and Machine Learning research is to: develop new intelligent methods to make rational decisions based on observations; learn from experience; and automatically extract knowledge and patterns from data. Thus, this paper explores the use a hybrid method to improve existing hybrids. We present a new method for a hybrid system, based on the spiral method.