Software Regression Testing Process Optimization For Agile Development

With the extensive application of agile software development process,the release cycle of software version is gradually shortened.In order to adapt to rapid iterative release,regression testing,as an important part of software testing,needs to be improved to adopt agile development process.However,at present,the actual regression test process used in industry does not consider the characteristics of rapid response of agile development process.Consequently,the test case execution efficiency is low,the fault detection rate is slow,and the versions with quality problems and delayed-release account for a high proportion of the total number of versions.In order to solve the problems,this thesis proposes the optimization of software regression testing process for agile development process.In order to connect with the agile development process,the regression test process is optimized by the ECRS analysis method and test case optimization technology.The "test case relation" was cancelled,the "round creation" and the "test case distribution" were merged.In addition,using test case optimization technology to sort and select test cases which improves the "test case execution".The test case sorting technology calculates the priority of the test cases and returns the sorting results by combining test cases with requirements and defects.The results of technical experiments proved that the sorted test case sequence can effectively increase the defect discovery rate.In the first round of regression testing,the average defect discovery rate increased by 26.7%?34.3%.The test case selection technology reduces the blindness of test cases selection by accurately selecting out test cases for interaction requirements.On average,each version can reduce the number of test cases for interaction requirements by 12%.In the second round of regression testing,using the test case ranking technology which combined with the selection technology,the defect discovery rate was increased by 35.3% on average,which greatly improved the efficiency of test execution.By comparing the cost,efficiency,and quality of the Smart Home APP before and after the optimization of the regression test process,it is proved that the optimized regression test process can effectively save costs,and improve the test quality,and reduce the delay rate of version release.In terms of cost,after the regression testing process is optimized,each version can save 1.53?2.88 person per day of testing workload,which can speed up the release of the version and save the cost of enterprise promotion,and save the cost of testing labor.In terms of quality,after the regression testing process was optimized,the proportion of versions with quality problems dropped by 16%.In terms of efficiency,after the regression test process was optimized,the version delay rate was reduced by 35%.Compared with 2020,the release cycle in2021 has been shortened by 1.53 days,but the total number of versions has increased by 11%,which better responds to the agile development.To sum up,the optimized regression test process saves the cost of promotion and test labor for the company,improves the quality and efficiency,better connects with agile development process,speeds up version iteration,and improves the company's market competitiveness.Therefore,the optimized regression test process has great promotion value and application prospects.