Space Mission Database From 1957 to 2022
Due to the recent news based on the crash of the Russia's first moon mission after nearly 50 years due to the crash with the surface of the moon, It piqued my interest to check on the pattern of the past space mission that have been done and since its almost independence day of Malaysia, so I picked from 1957 to 2022. This is my own personal project and I don't refer anyone or anything during the whole process except with the help of the tools mentioned.
01
Data collection
The first step is to find raw data of space mission from 1957 or before that until the recent one and gladly I found one from Maven Analytics database where it's exactly has everything that I need.
Link to raw data: Space mission dataset
02
Data cleaning
Now its time to clean the data. We all know that raw data often contains errors, inconsistencies, missing values, and outliers and to ensure that the data is accurate, reliable, and ready for further analysis or use in decision-making, we have to clean the data first. There are two tools that I used during the whole process of cleaning where I use excel for the web and Excel 2017.
-
First, convert the data from CSV into XSLX file.
-
Uniform all the column so that all the value can be read. I use "CTRL+SHIFT+RIGHT ARROW" and double click on any of the two column so that all data can be shown clearly.
-
And then I clicked on cell B2 and click Freeze Pane so that we can still see the column header if we slide horizontally and the company's name if we slide vertically.
-
Next, we click on any of the data and click filter and we can check if there's any blank data. We can see that there's a few blank box in Time and Price column. The data is too big hence I moved the excel into excel 2017. Instead of deleting it, by the process of "Go To+Special+(Blank)", "SHIFT+CTRL+DOWN ARROW" and insert all the value using "CTRL+Enter":
-
I put a NULL in the blank box of "Time" column due to that its irrelevent data to my analysis hence no need to delete the rows containing blank time.
-
I put a "0" in the blank box of "Price" to indicate that the company doesnt have any loss.
-
-
Now, lets check the spelling but apparently all the name is a special name hence not much we can change. in fact none tbh.
-
By custom sort, I put the order as the "Date+Increasing Order" followed by "Time+Increasing Order" for me to arrange it according to the first time space mission have been done to the latest according to the data because we want to analyse the timeline. What I noticed afterwards is that the date kinda hard to read due to it follows the format of MM/DD/YYYY so I number format it to follow DD/MM/YYYY for a proper way to read the data.
-
I insert another column to seclude a country from the address given as the analysis that I did will be focusing on the country as well hence I use the "Flash fill" to exclude the country from the address. After that, need to do some bit checking just to be sure that its being classified and sorted out correctly.
-
Lastly I format the worksheet into table to do pivot table of it
​
03
Data Validation
Next, we need to make sure that the data is reliable to be used for analysis or not. Specifically since we have create a new column known as "Country" We need to make sure that is it really owned by the company or simply just the location site of where the space mission been done. Hence, we'll use the help of Google and ChatGPT to help us found out if the representative country is the owner of the space mission at that current time. There are some cases we can observed here when we do the checking for example
​
-
There are company that currently being owned by other country in the present while it belongs to other country in the past. For example, Sea Launch that's actually a consortium of several countries before 2016 where it solely belongs to Russia.
-
There are private company as well such as SpaceX but we categorized them under their specific country because they somehow get funded as well.
-
Some company or mission that's being owned by several country but we only take the main shareholder.
04
Data Analysing
So, before we start visualising the data. We need to make sure that we know what is it we want to analyze. Hence, these are questions that I will tackle based on the recommendation from Maven Analytic itself which consisted of:
​
-
How have rocket launches trended across time? Has mission success rate increased?
-
Which countries have had the most successful space missions? Has it always been that way?
-
Which rocket has been used for the most space missions? Is it still active?
-
Are there any patterns you can notice with the launch locations?
​
Other than that, I add on another question on my own where
​
5. how much cost that each country have spent for every space mission?
​
Hence, this is the visualization that I will use to tackle all the problems through microsoft Power BI.
05
Data Visualization
There are a few iteration of my dashboard after getting a lot of insight and critique from Power Bi community in reddit. Honestly they are very helpful and the advice given are very constructive. I specifically get help from one specific individual who has help and guide me in completing this project. These are the timeline of what I have did for my dashboard
Draft: these are just draft that I did when I first started to try using Power BI. I learn a lot after these draft and making major improvement to my dashboard
1st iteration: I was planning to remove the map at first but somehow question 4 specifically asking for something to do with location. Hence, why instead of removing it, I put it on a drillthrough. There are also a few DAX involved involving the rate of success and the new price.
latest iteration: These are the latest update so far where I create a filters button that will cascade the filters and slicers. I remove the active rate of rocket because its irrelevant and add no value to the dashboard. I will do a new drillthrough with a better KPI of the map and the table as well but its still under progress.
There are still major improvement that has to be made hence it will take quite some time before I managed to complete everything and feel satisfied with the dashboard. This is my own personal work and not belong or copy from another third parties unless being mentioned. I will post it on my medium post as well when its being done. so look forward to it