The complexities of cyber physical systems development are mapping with controlling software in the modern engineering. Software testing are major key role in the modern engineering. The expected outcome from the test case suite is an important source for testing the software coding. Test results ensure the quality with the correctness of the developed software. The modification of the software source code has led to a change in the expected outcome in the regression testing. Validating a large number of test cases multiple times can be a high execution time and development cost. The methodology has been proposed for the selection of test paths during regression testing. The selected paths are covering new expected outcomes in the modified software program, thereby reducing the testing time. Regression test path selection methodology is based on Jaccard similarity mapping of control flow graph (CFG) and artificial bee colony (ABC) algorithm to find out the new test path which is not tested in the older version of the software programming. Our experiments show that the proposed approach identifies the new test path for test case generation during regression testing. The fault detection capability is high compared with structural novel test path selection methodology. Code coverage criteria for the modified software programs show high effectiveness with less test path selection.