Case study - Test Automation of a Mobile Application
0 likes1,120 views
The document discusses the test automation of a mobile distribution management system developed by Oak Systems, emphasizing the need for automation due to the extensive manual testing effort required for quarterly software releases. A hybrid test automation framework is proposed, supporting both mobile and browser interfaces while achieving a significant reduction in testing time and an increase in the number of automated tests. The implementation of this framework resulted in 95% of tests being automated, significantly enhancing testing efficiency and product confidence.
![Features of the Application
➢ Application is a DMS [Distribution Management System] with 3 levels of sale; viz. Primary, secondary,
tertiary. Has sales, stock modules.
➢ Mobile channel: It was a native app delivering solutions in Demand Chain Execution space covering –
Sales Influencing, Demand Fulfilment and Delivery Optimization, ensuring best in class Sales
Management, Order Taking, Direct Store Delivery, Merchandizing and Delivery functionalities using
optimization algorithms to provide fast and efficient supply.
➢ It Encompassed the space of Demand Chain Execution with solutions covering Sales force
effectiveness, Secondary sales execution and delivery optimization
➢ There were features available on android mobile which are on field data capture or transactions
which are real-time or offline (later integrated). Plan was on to make available on iOS
➢ For web based interface, only chrome browser was supported.](https://image.slidesharecdn.com/oaksys-casestudy-testautomationofamobileappplication-201105110953/85/Case-study-Test-Automation-of-a-Mobile-Application-3-320.jpg)
Case study - Test Automation of a Mobile Application
- 1. Test Automation of a Mobile Application Oak Systems Pvt. Ltd. oaksys.net
- 2. Application under consideration A Distribution Management System that is part of Enterprise Resource Planning (ERP) involving order processing ecosystem connecting Producer to Distributor and Retail stores in merchandise trade predominantly into apparels. Available with Web based and Mobile application front-ends.
- 3. Features of the Application ➢ Application is a DMS [Distribution Management System] with 3 levels of sale; viz. Primary, secondary, tertiary. Has sales, stock modules. ➢ Mobile channel: It was a native app delivering solutions in Demand Chain Execution space covering – Sales Influencing, Demand Fulfilment and Delivery Optimization, ensuring best in class Sales Management, Order Taking, Direct Store Delivery, Merchandizing and Delivery functionalities using optimization algorithms to provide fast and efficient supply. ➢ It Encompassed the space of Demand Chain Execution with solutions covering Sales force effectiveness, Secondary sales execution and delivery optimization ➢ There were features available on android mobile which are on field data capture or transactions which are real-time or offline (later integrated). Plan was on to make available on iOS ➢ For web based interface, only chrome browser was supported.
- 4. Trigger for Automation ➢ There were quarterly Planned Software releases to address defects and enhancements. Even if the release had few modifications, the Manual Testing of the release including the Regression Testing was consuming lot of effort and time. While meeting the release dates, the testing used to be curtailed resulting in release of partially tested product. ➢ Mobile application being newly built, compounded the problem of inability to produce clean releases ➢ End to End test cycle covering both mobile devices and Browser interfaces was growing (Around 600 test case per device) ➢ Manual Testing was prone to errors due to repeating same tests on Multiple Mobile OS and Multiple Browsers ➢ In addition, Multi-country (localization) tests were also increasing due to implementation of system across geographies
- 5. Our Solution Architecture ➢ Hybrid test automation framework to address both browser and mobile app interface seamlessly ➢ Dynamic selection of test candidates (application areas) for execution was facilitated ➢ Robust design for future enhancements (scalability) ➢ Use of open source tools to reduce TCO. ➢ Accordingly framework built in Appium (it supported both Android and iOS interface), Selenium (java based) using JavaScript and Database interface (instead of Spreadsheet) using python. Database interface ensured that the automation was faster ➢ To provide technical documentation for incorporation of new test cases into framework
- 6. Test Automation Engineering ➢ Several Automation methodologies like Page Object Model, Data Driven Approach were evaluated. Considering size of automation, Data Driven Approach was adopted ➢ Design to ensure adaptability to all leading Desktop OS browsers (Edge, Firefox, Chrome) and Mobile OS viz. iOS, Android ➢ Advised a disciplined approach of development to make uniform UI interface ➢ Redefining expected outcome of tests to suit automation needs ➢ Successful tests adapted for automation ➢ Database storage instead of traditional approach of Spreadsheet for storing test data. Due to database usage, speed of automation increased.
- 7. Outcome ➢ Hybrid test automation framework to address both browser and mobile app interface seamlessly ➢ Browsers Covered ▪ IE ▪ Firefox ▪ Chrome ➢ Mobile devices Covered ▪ iOS (v9.x) ▪ Android v6.x ▪ Android v8.x ➢ Country-specific (localization) tests ➢ Scalability to add new tests ➢ User Friendly interface to store test attributes in database
- 8. Return on Investment - 1 ➢ As Freeware tools were used for automation, there was no investment on automation tools ➢ The framework provided to easily add New feature Tests needing no automation enhancements ➢ 95% of tests automated as against the industry std recommendation of 85-90% possible automatable tests 95% 85% 5% 15% Actual achieved Industry std %geofTestsAutomated
- 9. Return on Investment - 2 ➢ Testing effort in each release is drastically reduced. One-time investment made for test automation will be recovered in testing 2-3 releases, after which, it is free. Automated Test EffortManual Test Effort
- 10. Return on Investment - 3 ➢ Every release is 100% regression tested thus improving confidence in the product and reducing instances of production defects ➢ Faster execution of tests and nightly execution (more than 70% reduction in Testing schedule) enabled more time for developers to fix the defects and ensured timely market release 0 10 20 30 40 50 60 70 80 90 100 Automation Testing Manual Testing Development Testing Fix & Release Product Development Product Development Testing (regression + new features) Testing (regression + new features) Elapsed Time (Calendar Days) Very less/no window to fix the defects within release date Completely Tested and bug free Product delivered ahead of time.
- 11. Thank you! Oak Systems Pvt. Ltd. #27, NGEF Lane, Off 100 Feet Road, Indira Nagar, Bengaluru – 560 038, INDIA Tel : +91-80-4126 7681 Email: [email protected] URL: www.oaksys.net We add value through engagement