Difference between revisions of "ANLY482 AY2017-18T2 Group07"
| Line 65: | Line 65: | ||
</li> | </li> | ||
<li> Operational Performance | <li> Operational Performance | ||
| − | <ul><li>Customers want to refine contract volumes to allow higher accuracy for DHL</li> | + | <ul> |
| + | <li>Customers want to refine contract volumes to allow higher accuracy for DHL</li> | ||
<li>Product managers realize there is a correlation in out the door(OTD) performance, volume, and surcharges, but no effective way to visualize them without multiple monitors running different dashboards.</li> | <li>Product managers realize there is a correlation in out the door(OTD) performance, volume, and surcharges, but no effective way to visualize them without multiple monitors running different dashboards.</li> | ||
<li>CSI team realizes a similar correlation and want to derive insights from these factors by putting them across a single page.</li> | <li>CSI team realizes a similar correlation and want to derive insights from these factors by putting them across a single page.</li> | ||
| Line 79: | Line 80: | ||
<p> | <p> | ||
<font color="#212121" face= "Franklin Gothic Book" size=4px> | <font color="#212121" face= "Franklin Gothic Book" size=4px> | ||
| − | + | Our team wants to help the sponsor make better sense of their data. During this preliminary stage, we intend to achieve the following: | |
<ol> | <ol> | ||
| − | <li> | + | <li>Data cleanup: |
| − | <li> | + | <ul> |
| − | <li> | + | <li>Inconsistencies in the data, for example- a text input in a numeric column</li> |
| − | <li> | + | <li>Missing values in the data</li> |
| + | </ul> | ||
| + | </li> | ||
| + | <li>Exploratory Data Analysis: | ||
| + | <ul> | ||
| + | <li>Basic exploratory analysis to check skews in the data and identify general trends</li> | ||
| + | <li>Identify the bottlenecks in the shipment journey affecting the operational performance</li> | ||
| + | <li>Analyze the shipment patterns and trends for lane-wise pairs in the existing dataset</li> | ||
| + | </ul> | ||
| + | </li> | ||
</ol> | </ol> | ||
| − | + | After the initial round of EDA, our team concluded that to help our sponsor in a sustainable manner, it’ll be essential to develop a visualization tool that encapsulates all key datasets. | |
</font> | </font> | ||
</p> | </p> | ||
Revision as of 17:04, 25 February 2018
Team Data Heavy Legends(DHL) wants to help its sponsor provide its service in the most efficient manner possible by making sense of the data they currently have. Through our efforts we hope to satisfy all stakeholders involved equally.
Motivation
Different stakeholder groups associated with our client, including their customers, Product Managers, and their GAM team alike have voiced concerns with the current data visualization platform:
- Peak Season
- Customers want to plan their delivery schedules ahead to build their business case.
- Product Managers must exercise prioritization and notify affected customers of potential delays.
- CSI team wants to prepare customers with data before the beginning of peak season, to help them build accurate business cases.
- Operational Performance
- Customers want to refine contract volumes to allow higher accuracy for DHL
- Product managers realize there is a correlation in out the door(OTD) performance, volume, and surcharges, but no effective way to visualize them without multiple monitors running different dashboards.
- CSI team realizes a similar correlation and want to derive insights from these factors by putting them across a single page.
In line with our client's digitization principles, the GAM team wants to reduce manual deductions by combining operational performance factors into a single page dashboard. Thus, through the use of filters, the dashboard would provide the GAM team with insights and forecasts into future customer demands, and help them predict lane wise traffic.
Objectives
Our team wants to help the sponsor make better sense of their data. During this preliminary stage, we intend to achieve the following:
- Data cleanup:
- Inconsistencies in the data, for example- a text input in a numeric column
- Missing values in the data
- Exploratory Data Analysis:
- Basic exploratory analysis to check skews in the data and identify general trends
- Identify the bottlenecks in the shipment journey affecting the operational performance
- Analyze the shipment patterns and trends for lane-wise pairs in the existing dataset
After the initial round of EDA, our team concluded that to help our sponsor in a sustainable manner, it’ll be essential to develop a visualization tool that encapsulates all key datasets.
