Presented by:

Professor:

Subject:

Date:

Top Slice Case

Introduction

The following is a case of the Top Slice Company that deals with making golf balls and producing oversized drivers. The company makes three golf models namely the Bombers, the Hool King, and the Sir Slice-A-Lot.

The problem:

The production manager Jacob Lee faces a big challenge because of the low production level experienced in the company. The company did not achieve the expected production of 2,700 drivers. In order to determine the best strategy that can make the company manage producing 2,700 drivers each month, Lee seeks advice through sales forecast models. Organizations use many forecasting models to suit their different business needs.

Relevant information to the problem

The Top Slice Case chooses between four forecasting models. These are the moving average model, the weighted moving average model, trend analysis model, and the seasonal model. Selecting the best forecasting model to use is a big challenge especially for Top Slice Company that has a large product line. The following discussion will help Lee select the best modeling technique and show the best time for the company to use the expanded work cell.

Solution

Question one

Which models technique best suits Jacob Lee?

The case made use of four modeling techniques. The mean absolute deviation for each model was calculated and used to determine the best out of the four models used. The mean absolute deviation (MD) forms the best statistical tool for determining most effective forecasting modeling technique. The MAD shows the mean differences between the predicted values and the actual values in a given time series. A low MAD shows that the model fits the time-series data being analyzed while a high MAD shows a poor forecast for a given model. Table 1 shows the MAD values for the four models used in analyzing Top Slice Company for a period of two years.

  MEAN ABSOLUTE DEVIATION

   

Forecast model Bomber Hook king Slice-A-Lot

3-point moving average 76.08 29.68 13.86

5-point moving average 69.03 27.18 13.23

3-point weighted moving average 62.23 23.74 12.69

5-point weighted moving average 65.3 5.94 18.19

Trend analysis 97.81 35.4 14.09

4-season model 0.08 0.002 0.106

8-season model 0.07 4.475 0.086

Table 1: Mean Absolute Deviation (MAD) values for four modeling techniques used in the Top Slice Company case analysis

The yellow highlighted sections indicate show the models with the lowest MAD values. From the table, the season model recorded the lowest MAD values and was selected as the best modeling technique. Both the 4-season model and the 8-season model recorded lower values compared to the other three models. The three company models, Bomber, Hook King, and Slice-A-Lot recorded MAD of 0.07, 0.002, and 0.086 respectively.

Why was the season model selected?

In statistics, error measurements play an essential role in benchmarking a forecasting process, tracking forecast accuracy, and monitoring any exceptions. The MAD is an error measurement tool used in statistics that measures the size of the error using units. Forecast accuracy is 100%, which is the accuracy is ranging from 0-1. From the results obtained, the season model formed the only modeling technique that recorded accuracies of between 0 and 1, hence, selected as the best modeling technique for Jacob Lee to adopt.

The assumption behind the model

The biggest assumption made behind this model is that the future can be forecasted to some degree of accuracy based on the past because the future environment and market are similar to what was available in the past.

Question two

When will Top-Slice need to have expanded work cell up and running?

From question one, the best modeling technique that suits the company is the seasonal model. The model will be used to forecast sales for the three categories until their production total reaches 2,700 drivers. Table 2 shows the sales forecast working.

Month Bomber Hook king Slice-A-Lot  Total

Mar-06 1720 490 399 2609

April 1840 491 433 2765

May 1969 492 471 2932

June 2107 493 511 3111

July 2255 494 555 3303

August 2412 495 603 3510

September 2581 496 655 3732

Table 2: Sales forecast for the coming six months of 2006

From table 2, Top-Slice Company would be ready work cell up and running in April 2006. Jacob Lee claimed that the available manufacturing work cell in the company was capable of producing 2,700 drivers in one month. The sales forecast calculated from the seasonal index shown in table 1 has seen the company approaching the value of 2,700 drivers in April that exceeds the targeted value by 765 pieces.

What are the implications for when Jacob should start the expansion effort?

The Top-Slice case clearly indicates that it takes at least three months to strategize and implement an expanded work cell. The results show that Jacob needs to expand the work cell in April. Jacob should start the implementation process in January 2006.

Conclusion

The availability of an experienced professional in business statistics helps an organization determine its sales forecast in an efficient manner. Different methods of business forecasting are available today, but each organization chooses a method that suits its needs. Quantitative methods of business forecasting like the ones used in this case are the best because they involve calculations using actual sales figures. The above analysis play a critical role in helping Jacob plan and implement the company’s cell expansion plan. Out of the four modeling techniques used, the Seasonal model worked out to be the best for forecasting sales at Top-Slice Company. In addition, it was realized that the company would expand the work cell up in April 2006, and Jacob would start planning and implementing the expansion program in January 2006.