Difference between revisions of "Improved Decisions for Ocean FreightsAnalysis"
(5 intermediate revisions by 2 users not shown) | |||
Line 148: | Line 148: | ||
[[Improved_Decisions_for_Ocean_Freights|<span style="color:#ffffff"><strong>Boxplots</strong></span>]] | [[Improved_Decisions_for_Ocean_Freights|<span style="color:#ffffff"><strong>Boxplots</strong></span>]] | ||
|- | |- | ||
− | |<br><center>[[File:Boxplots.PNG| | + | |<br><center>[[File:Boxplots.PNG|400px]]</center> |
We also provide box plots to complement the cumulative distribution graphs so that the distribution can be better understood. | We also provide box plots to complement the cumulative distribution graphs so that the distribution can be better understood. | ||
Line 160: | Line 160: | ||
[[Improved_Decisions_for_Ocean_Freights|<span style="color:#ffffff"><strong>Cost Treemap</strong></span>]] | [[Improved_Decisions_for_Ocean_Freights|<span style="color:#ffffff"><strong>Cost Treemap</strong></span>]] | ||
|- | |- | ||
− | |<br><center>[[File: | + | |<br><center>[[File:Cost1.PNG|800px]]</center> |
− | + | In this visualization, the different colours represent the different trade routes and the size of the boxes indicate the total cost of all the transactions in that trade route for a specific company. | |
Line 172: | Line 172: | ||
[[Improved_Decisions_for_Ocean_Freights|<span style="color:#ffffff"><strong>FCL/LCL Cost Comparison</strong></span>]] | [[Improved_Decisions_for_Ocean_Freights|<span style="color:#ffffff"><strong>FCL/LCL Cost Comparison</strong></span>]] | ||
|- | |- | ||
− | |<br><center>[[File: | + | |<br><center>[[File:Costwhatif.PNG|800px]]</center> |
+ | |||
+ | In order to convince companies to switch from FCL to LCL, one of the factors to consider is cost. For FCL transactions, the cost is calculated based on the number of TEUs purchased by the customer. On the other hand, the cost of LCL transactions is determined based on the volume. As such, it is important to determine the breakeven volume where LCL starts becoming more expensive than FCL. Unfortunately, we do not have cost data from DHL. To circumvent this issue, we allow users to input the cost per cubic metre for LCL transactions, and the cost per TEU for FCL transactions, by themselves before producing the analysis. | ||
− | |||
Line 186: | Line 187: | ||
|<br><center>[[File:DashboardDesign.png|800px]]</center> | |<br><center>[[File:DashboardDesign.png|800px]]</center> | ||
− | + | This is the screenshot of our dashboard for the interims. During the interims, we imagined the dashboard to be in the form of a set of questions where each page will provide a suggestion for the answer to each question, then there would be three pages to the dashboard: | |
:<li>Customer Profiling | :<li>Customer Profiling | ||
:<li>Trade Lanes Analysis | :<li>Trade Lanes Analysis | ||
:<li>What-If Analysis | :<li>What-If Analysis | ||
On each page of the dashboard, there will be a series of visualisations which will each provide a unique insight that are interrelated to provide the answer to the question. The above diagram depicts how we visualize how the various insights (Customer Profile) are interrelated. | On each page of the dashboard, there will be a series of visualisations which will each provide a unique insight that are interrelated to provide the answer to the question. The above diagram depicts how we visualize how the various insights (Customer Profile) are interrelated. | ||
+ | |||
+ | Our final dashboard still has the same philosophy, just that we have removed the What-If Analysis page as we felt it would not provide effective insights. Here are screenshots of our final dashboards: <br/> | ||
+ | <br><center>[[File:Dashboard1.PNG|600px]]</center> | ||
+ | <br><center>[[File:Dashboard22.PNG|600px]]</center> | ||
+ | <br><center>[[File:Dashboard3.PNG|600px]]</center> | ||
+ | |} | ||
+ | |||
+ | <br /> | ||
+ | {| style="background-color:#ffffff; width:80%; font-family:Century Gothic; font-size:15px; margin: 3px auto 0 auto;" | | ||
+ | | style="background-color:#006600; ; color:#ffffff; text-align: center; border-top:solid #ffffff; border-bottom:solid #ffffff; width:50%; " | | ||
+ | [[Improved_Decisions_for_Ocean_Freights|<span style="color:#ffffff"><strong>Extra visualization</strong></span>]] | ||
+ | |- | ||
+ | |<br><center>[[File:Extra1.jpg|600px]]</center> | ||
+ | |||
+ | After our final presentation, we had some feedback from our client and we made this visualization to address the feedback. In this visualization, different bubbles indicate different ports of load and the size of the bubbles indicate the amount of carbon emitted from transactions originating from that port. In addition, users can also look at the average percentage utilizations of transactions originating from that port as well as the total number of transactions. The visualization is filterable by percent groups. In the above example, only ports with utilization rates of 0 to 25% are shown. | ||
|} | |} |
Latest revision as of 20:05, 16 November 2015
No clear relationship between shipper country and utilization rate |
![]() We see from the graph on the left that shipments originating from Chile, Singapore and the United States have the lowest utilization rates. However, we see from the graph on the right that shipments originating from Fiji, Italy and Estonia have the lowest utilization rates. As such, we are unable to see a clear relationship between the shipper country and utilization rates. |
No clear relationship between consignee country and utilization rate |
![]() With the same X-axis and Y-axis, that is Consignee Country and Average Percentage Utilization, we realise that are no similar trends between the 2 graphs of different industries (auto industry and engineering industry).
|
Underutilization could possibly be due to danger of goods involved |
![]() We attempted to analyse if the danger level of the goods affected the choice of container.
We can see from the graph above that 5 out of 6 industries underutilize the containers when dangerous goods are involved. However, we also realized that there are only 142 records of dangerous goods available, as compared to 82,649 records of non-dangerous goods. Due to the vast difference in numbers, we are not able to say with certainty that the danger level of the goods affects the ultimate container choice. |
Breakdown of average utilization by Industries for FCL and LCL |
![]()
Out of the 6 industries, it becomes apparent that, in the order of lowest utilization of FCL are:
As such, we would suggest focusing on Engineering companies first. |
We also provide box plots to complement the cumulative distribution graphs so that the distribution can be better understood.
|
In this visualization, the different colours represent the different trade routes and the size of the boxes indicate the total cost of all the transactions in that trade route for a specific company.
|