To early discover faults in source code, test case ordering has to be properly chosen. To this aim test prioritization techniques can be used. Several of these techniques leave out the execution cost of test cases and exploit a single objective function (e.g., code or requirements coverage). In this paper, we present a multi-objective test prioritization technique that determines sequences of test cases that maximize the number of discovered faults that are both technical and business critical. The technique uses the information related to the code and requirements coverage, as well as the execution cost of each test case. The approach also uses recovered trace ability links among source code and system requirements via the Latent Semantic Indexing technique. We evaluated our proposal against both a random prioritization technique and two single-objective prioritization techniques on two Java applications. The results indicate that our proposal outperforms the baseline techniques and that additional improvements are still possible.