Project Balto – Some Key Lessons in Test Planning
Here are the top four responses from our readers on Project Balto testing challenge. Mike will publish a detailed article on some key lessons learned during the Project Balto. Stay tuned for our February 2016 edition.
In no particular order. Prizes are being sent to participants.
Response 1: (Pramod Gupta)
No Testing is very risky for a critical space mission like Project Balto. Below are few ideas to improve the success of the Project Balto:
- Stage Assembly
- Revisit the Features of each stage and select minimum features / must have features required to make this mission successful – If this will work by reducing few features we can cut down the Stage assembly time lines without risking mission of the Project Balto. If the all the features listed is must have then go to #2
- Stage Testing – Execute Risk based Testing only
- Stage 1 has been used and many times so have less risk of reusing it , a minimal test of separation charges will do
- Stage 2 also has been used more than 50 times hence very reliable and will be good with no stage testing
- Stage 3 – Do a risk based testing of the features critical for the mission
- Freight train to Canaveral
- Bring the ready stages to Rocket Integration location as soon they are ready rather than all coming at the same day.
- Some of the stage can come earlier than other to the integration location
- Rocket Integration
- Need not to wait for all 3 stages arrived to start integration and can start stage 1 and stage ( Stage 1 and Stage 2 are reusable hence probably does not need as much time as Stage 3)
- Rocket Testing can be done in parallel to the residual rocket integration activity
- Plan a stage integration test rather than big bang approach
- Do a top down or bottom up stage wise integration test
- By executing any parallel activity where ever possible / feasible gain some time which can be used for risk based testing
- Following the updated schedule we still can afford to invest at least 1 day can be dedicated to stage testing and 2 days to integration testing will make ensure the success of the mission
- By working round the clock means we can reduce the duration by 1/3 in the revised schedule duration is just cut down by half, so there is other window of opportunity to in terms of timing.
Response 2: (Sthitipragyan Mohapatra)
Yes for sure we should not skip the entire testing phase in this case. If any of the travelling time cannot be reduced, we have to revise the time of ‘Stage assembly’ and ‘Rocket Integration’.
As per my knowledge stage testing can be comparable to functional testing for each part of a particular stage.
And Rocket testing can be comparable as combined Integration and System testing.
System testing very important before launching, so we have to fasten the rocket integration process to get some time for System testing (5 days,3 days). We should prioritize the test scenarios based on their criticality and try to cover the most required and critical features first. As the Liftoff stage is getting attached with this unit for the first time to get to the moon, this part need more testing than the other two.
Similarly we have to reduce the time taken for stage assembly so that some of the major individual parts of each stage that need to be functioning for current requirement should be tested once. If possible can made it to 10 days and 5 days.
I don’t have much knowledge on rocket and its functioning parts. But if any of the feature (camera) is having low priority of being used in current requirement, testing for the same could be skipped. Also based on the rescue or resupply operation selected we can revise the testing efforts.
Response 3: (Ayush Kapoor)
Some of the things which we can do:-
This is the best example to Use RAD – Rapid Application Development
- Not ideally waiting for the “Freight Train to reach Canaveral” we should Continue Testing and Bug Fixing so the we have 2 More Days.
- We have 30 Days in Hand to save the crew and as shown in the last Revised Schedule is for 28 days we can still 2 daysleft which we can assign them 1 day for “Stage Testing” and 1 Day for Rocket Testing.
- We can also not keep waiting for the Stage to be prepared fully we should start testing it from the “Stage Assembly phase” so we can use around 2-3 days by simultaneously testing the Module
- Again simultaneously with Rocket Integration we should continue the Testing.
So the Revised Schedule:-
Response 4: (Ishani Perera)
Problem Analysis and Assumptions
- Maximum survival days on moon — 30 – 31 days
- Final project timeline — 28 days
- No allocated testing time.
- Building a new rocket not reusing any stages.
- Recovering stages are not a priority/ important.
- All staff working to the maximum capacity.
- Co-locate testers with the supplier’s assembly specialists and the integration engineers.
- Communicate with the supplier engineers and centre engineers regarding the identified components and get approval for each stage to be completed only with the most important/ required components.
- For each of the three stages, identify the components/ features that have no critically importance for the success of the mission.
- Stage 1 – Recovery parachute, Recovery beacon, Re-entry shields
- Stage 2 – Recovery parachute, Recovery beacon, Re-entry shields, Black box monitor
- Stage 3 – Black box monitor, Radio
- Since there is no testing time allocated in the timeline, take a more agile approach for testing by testing while building;
- Testing in each stage during the assembly stages.
- Test during rocket integration stage.
- Exploratory testing and defect retesting to be carried out after rocket integration, as there are additional 2 -3 days left from the estimated delivery days.
https://www.testingcircus.com/project-balto-some-key-lessons-in-test-planning/https://i1.wp.com/www.testingcircus.com/wp-content/uploads/Project-Balto-024.jpg?fit=620%2C388&ssl=1https://i1.wp.com/www.testingcircus.com/wp-content/uploads/Project-Balto-024.jpg?resize=150%2C150&ssl=1ArticlesProject BaltoHere are the top four responses from our readers on Project Balto testing challenge. Mike will publish a detailed article on some key lessons learned during the Project Balto. Stay tuned for our February 2016 edition. In no particular order. Prizes are being sent to participants. Response 1: (Pramod Gupta) No Testing...Ajoy Kumar SinghaAjoy Kumar Singha[email protected]AdministratorAjoy is the founder and editor of Testing Circus magazine which is read and subscribed by thousands of professional testers around the world. He is associated with various testing forums such as NCR Testers Monthly Meet as a founding member. Follow Ajoy on Twitter.Testing Circus
Ajoy Kumar Singha
Latest posts by Ajoy Kumar Singha (see all)
- Project Balto – Some Key Lessons in Test Planning - February 25, 2016
- Testing Circus – December 2015 Edition - January 10, 2016
- 10 Million USD Research Project to Guarantee Bug-Free Software - January 8, 2016