The Mobile Phone Material Safety Stock Forecast Based On Genetic Algorithm And Neural Network

Mobile communication has become one important part of people's daily life, With China's mobile communications'fast development, more and more people own telephone,which brings great business opportunities for mobile phone manufacture.At present, there are so many mobile phone production enterprises that many of which have to facing fiercely competitive,if they want to keep ahead in the competition,they have to keep their production be differentiation, high quality,good service , advanced technology and what's more the cost control is also important. The key factor of cost control is to control raw material inventory. Keeping mass of raw material inventory can effectively prevent production line from stopping and avoiding the loss of order delay,but at the same time inventory takes up enterprise's capital, produce maintaining cost of inventory,and its also have the risk of overstocked loss.Therefore, the enterprise in the business, should avoid inventory be in shortage, and also should avoid inventory excess, reduce unnecessary inventory costs.Maintaining reasonable inventory, improving safety stock forecast level, which is important to cost reduction,enhance the competitiveness of enterprises.Mobile phone is high-tech products in electronic industry, its function is advanced,its structure is complex, and its components have many kinds.To produce an ordinary mobile phone need the ectronic material of resistor, capacitor, crystals, IC, flash and etc..Its also need structure material like keyboard, LCD, battery and ect.A mobile phone usually needs hundreds of kinds material,a smartphone need even more. Mobile phone manufacturers usually produce several different types of cell phones at the same time, different types of cell phones usually need different type of raw materials,which will brings geometric progression increase of raw material inventory in the enterprise.So many kinds of inventor brings great difficulties for inventory management, safety stock forecast, the traditional safety stock forecast method is hard to meet the requirements,so in this paper neural network will be used in safety stock forecast.The relationship between the factors that affect the safety stock is complex and difficult to forcast, and between these factors and the decision results exist a complicated nonlinear relation. Neural network has unique and incomparable advantages in dealing with nonlinear question,so in this paper we use neural network to forcast the safe stock. We select five key factors that influence the safety stock as input data, build a BP neural network model to forecast the safety stock. Because BP neural network has the shortcoming of easily falling into the local minimum, we use genetic algorithm to optimize the BP neural network model,which can improve the BP neural network search ability, so we build the BP neural network model based on the genetic algorithm and experiments show the optimized BP network model has high precision and stability.Finally we use Using Matlab and VB to programme developed safety stock prediction system which has simple interface and easily to operate.