Automated unit test generation has been extensively studied in the literature in recent years. Previous studies on open source systems have shown that test generation tools are quite effective at detecting faults, but how effective and applicable are they in an industrial application? In this thesis, this question is investigated in two phases. In the first phase, I empirically investigate the effectiveness and applicability of existing automated unit test generation tools and techniques in an industrial financial application known as LifeCalc which is a life insurance products calculator engine owned by SEB Life & Pension Holding AB Riga Branch.

In the second phase, I focus more on the software characteristics of financial application domain. In this domain, many legacy applications exist as a collection of formulas implemented in spreadsheets. These legacy code, at some point, will be migrated to more modern development environment. However, migration of such code to a full-edged system is an error-prone process. While small differences in the outputs of numerical calculations produced by the two artifacts are tolerable, large discrepancies could have serious financial implications. Therefore, in this phase, I introduce a novel specialized search-based unit test generation technique that seeks to uncover the deviation failures in the migrated code automatically.