Neural Forecasting Network for the Market of Pleione Formosana Hayata Orchid

Abstract

Pleione Formosana Hayata Orchid is one of Taiwan’s native plants. Growing high in the mountains at an elevation of 1500-2000 meters, it requires a temperature range of 15-20 degrees Celsius. This perennial is a member of the family orchidaceous. It has one bulb and only one leaf. It is sold in bulb form, and it blooms before its leaf forms. At harvesting time, nursery personnel have always had to invest much capital to stock Pleione Formosana bulbs for the traditional orchid industry. However, the price of Pleione Formosana bulbs changes daily based on market supply and demand. This fluctuation makes it difficult to know how many bulbs to stock at any given time. However, if information technology could be used to assist operating personnel to forecast the demand for the flower in the near future, they can buy at a low price and achieve the objective of short-term stock, according to short-term demands, without misjudging the amount to be bought. Thus, they not only can increase their profit, but also enable customers to get fresh bulbs at a low price, thereby assisting them to reduce material costs. This research presents a market demand forecasting system for the Pleione Formosana Hayata Orchid product to assist traditional market personnel to forecast customer demand in the near future. The back propagation neural network algorithm will be used in the Pleione Formosana Hayata Orchid product market demand forecasting system so that the order demands in the future can be forecast on the basis of information on existing orders.