Forecasting techniques (total of 10 marks for this case study)

Hoa Phat furniture (HPF) becomes one of the Vietnamese leading company with significant growth in furniture production. The company’s business strategy has been consistently maintained good quality and on-time delivery. The main product line of the company: High-quality office cabinet (OFC01). Hardware, fixtures and accessories can be produced and supplied by local manufacturers, by traders and from foreign sources. Recently, during expansion to fully covering the entire market, the company faces several challenges in maintain quality as well as forecast issues. In addition, the Covid-19 disruptions influence significantly on the forecast and delivery schedules, which eventually can affect company orders and financial positions.

The company has had steady growth in the last 12 months. Given the Covid19 situation and uncertainty in regional and global trade tensions, the Operations manager become more cautious in next forecast. He asked the company to provide accumulated demand data shown in Table 1 for OFC01for the past 12 months.

Table1 HPF Demandin the last12 periods

The manager of the company wants to be certain enough OFC01 are available to complete orders promptly and they have adequate inventory in stock. During the last meeting, a number of forecast techniques have been discussed:

1. Simple moving average n = 4

2. Weighted moving average with n=4 with the following weights from most recent periods: 0.4, 0.3, 0.2, 0.1

3. Exponential Smoothing with alpha = 0.2

4. Linear trend

The manager would like to ascertain the best tools that he and his team can effectively manage the production processes. Among the above forecasting methods, which method should be used to forecast the demand for this product? Why?

The manager specifically interested with Exponential smoothing technique; however, he was not sure how to select the alpha coefficient. He learned that the most commonly used values of alpha are in the range of 0.1 to 0.5. His friends from RMIT recommends he could test these values of alpha in the range of 0.3 to 0.5. He wonders if there is a way that he can find an optimal value of alpha? Please provide your recommendations to his concerns.