Difference between revisions of "G7 Methodology"

From Analytics Practicum
Jump to navigation Jump to search
Line 47: Line 47:
 
<!--Content Begins-->
 
<!--Content Begins-->
  
[[Image:DHL_Methodology.png|center|2050x332px]]
+
[[Image:DHL_Methodology.png|center]]
  
 
<font color="#212121" face= "Franklin Gothic Book" size=4px>
 
<font color="#212121" face= "Franklin Gothic Book" size=4px>

Revision as of 13:19, 15 April 2018

DHL Common Banner.png

HOME

 

PROJECT OVERVIEW

 

PROJECT FINDINGS

 

PROJECT MANAGEMENT

 

ABOUT US

 

PRACTICUM HOMEPAGE

 

DHL Methodology.png

Figure 2 shows the Methodology applied by the team. The dataset contains shipment records of each BU of Company X’s largest client for the years of 2015, 2016, and 2017. The team began data preparation by creating an appended file and calculating columns representative of the four phases of the shipment process. Discrepancies found from this process were highlighted to Company X. Following this, an Exploratory Analysis was conducted – this helped the team detect anomalies and outliers in the data. To determine statistical significance between the parameters being tested, a further confirmatory analysis was performed using methods such as Phi and Cramer’s test, Kruskal-Wallis test, and Mann-Whitney test. Following this, a dashboard was created to draw insights using visualization tools such as Mosaic Plots, and Box Plots. The following methodology allowed the team to formulate recommendations for Company X’s GAMs.