Developing Model for Customer Demand Forecasting and Inventory Management by The Single Exponential Smoothing Algorithm
The objective in this research is to develop modeling for customer demand forecasting and inventory
management by exponential smoothing algorithm. The perishable goods are tested in this modeling. Because of Thailand has
many agricultural products and makes profit from those products but control of this type of goods is also problematic. Then
the forecasting model is created to apply in planning the quantity of goods to warehouse for the lowest volume of storage.
Sale data volumes of a wholesale of pork chop at Bungkan province are recorded in this research. The training data sets are
the volumes of pork chop in the past 350 days for the forecasting model. The experiment results in this research found that
the single exponential smoothing algorithm is higher that the adaptive-response-rate single exponential smoothing algorithm
and the Holt’s two-parameter linear exponential smoothing algorithm. Then the single exponential smoothing (SES)
algorithm is selected and is applied to use for developing in the perishable goods (PG) model. The advantage of this research
is making profit more than the old system that this mode that this model gives equal 120,320 Thai Baht or 3,758.82 USD
while the old system is 99,950 Thai Baht or 3,122 USD. Moreover, the PG model can decrease inventory balance more than
the old system that is 220 kilograms approximately.
Keywords - Single Exponential Smoothing Algorithm, Adaptive-Response-Rate Single Exponential Smoothing Algorithm,
And Holt’s Two-Parameter Linear Exponential Smoothing Algorithm.