G7 Methodology

HOME

PROJECT OVERVIEW

PROJECT FINDINGS

PROJECT MANAGEMENT

DOCUMENTATION

ABOUT US

PRACTICUM HOMEPAGE

 

Sponsor Data

Team Methodology

Work Scope

Project Timeline

Methodology

Methodology For this project, we will use a 3 Phase approach involving data cleaning and preparation, creating actual modules and sensitivity analysis.
Our team is required by the sponsor to use Power BI as the visualization tool to maintain consistency across the company. Since we have not signed the NDA yet, so the sponsor has only provided us with the field names along with a data dictionary rather than the actual data, so we can only share our initial thought process on the methodology which will be involved to create each module.

1. Lane Seasonality: By analysing measures like number of shipments, volumetric weight and cost of shipments, we shall try to spot seasonal trends for each lane per account. This will also include analysing the main carriers (airlines) used per lane. A dynamic dashboard will show these trends with options to filter by region, lane and account.
2. Density Trend Analysis: The airlines use a pricing structure in which the price of shipping is determined by both the actual weight and the volume of the shipment. Our sponsor has provided us with the methodology to calculate this chargeable volumetric weight as accepted in the industry. We will use this methodology to categorize the shipments and spot density trends, so we can minimize the shipping costs by having a mix of dense and non-dense shipments.
3. OTP Latency: The shipment data contains the timestamps for each stage in the airfreight operational flow from origin to destination. Based on these we can determine the latency in delivering the shipments for different service categories like ‘Airport to Airport’ (ATA), ‘Door to Airport’(DTA), etc. We further hypothesis that this latency could be due to numerous factors like air carrier, density of shipment, number of shipments per lane, etc. So, we can use statistical techniques like multiple regression analysis to determine these factors.
4. Capacity Planner: Based on the seasonality and trends found in the earlier modules, we will try to forecast the capacity required each month. We will use several forecasting methods like Moving Average, Multiple Linear regression, etc and choose the forecast with the lowest Standard Error. We will make the model dynamic so that it improves with time as more data is added to it.

Overall, Project STP+ will map the road towards efficient collaboration, and drive towards greater business growth.