A Cat Corp Final Paper

...

The data that the teams are given in the case study appears to be units used/sold and manufactured and this is good data to import into Excel. As far as the team could tell the only variable that is present is the year, quarter, or month that the units were sold/produced in, so this would mean that the team wouldn’t need to use any kind of more sophisticated software to do the analysis with. All of the data are numeric, which means that the data are quantitative and not qualitative. Quantitative data can be measured and is in this example, sales in each identified period is measured. Excel is used when one has quantitative data. It would not be as useful if the team had qualitative data to work with. The team would then need a different software package to complete the analysis.

We used descriptive statistics from the Excel Tool Pak to analyze the data. The team was given five years of data to use for the analysis. The team is using the descriptive statistics because we are interested in the mean. We used the measurement of simple moving average, or SMA. It has been shown to be an effective trend indicator in the historical analysis of number prediction (Maverick, 2014). We have used both of these tools to help the company to determine how many transformers they should be purchasing, as well as the quality department who could create control charts and intervals to help alleviate bottlenecks and ensure that they are producing on schedule to meet the quota that they could enact based on the analysis that we are presenting. Simple Moving Average is when one takes the last number in each period and then divides that by the number of periods. This is an easy way to see the trends in the data. Descriptive statistics are used in statistical analysis to summarize the data by looking at measurements such as mean, standard error, mode, and standard deviation. This method will allow for the most reliable data because we are able to look at a full five years of data and how they relate to each other.

Next we are analyzing the data we received from the company. Our purchasing department told us that sales increased 10% each year and we needed to take that into account when we forecast the future needs. We have to use the statistical analysis to reach a decision to our problem of how many transformers we should be purchasing to produce enough of our products to make our quota that we establish. Our hypothesis is that the mean of the transformers needed is over 1000 units per month. Our null hypothesis is that we need exactly 1000 units per month, and the alternative hypothesis is that we need less than 1000 units per month. We also have to look at the info we were given about the 10% lift each year. In this case, the data should show that units needed would increase by 10% each year. We could choose to analyze all the data we have and predict beyond the data we have, or use the data we have and split it. This way we could see if our predictions are true with actual data. We are going to calculate a 3 year moving average using 2006-2008 data to predict the data results for 2009, and then 2007-2009 to calculate 2010. This should give us the statistics to see if our hypothesis is true.

EXH 1

2006

2007

2008

2009F

2010F

2009A

2010A

January

779

845

857

827

843

917

887

February

802

739

881

807

809

956

892

March

818

871

937

875

894

1001

997

April

888

927

1159

991

1026

1142

1118

May

898

1133

1072

1034

1080

1276

1197

June

902

1124

1246

1091

1154

1356

1256

July

916

1056

1198

1057

1104

1288

1202

August

708

889

922

840

884

1082

1170

September

695

857

798

...