Evolutionary algorithms are among the most common techniques developed to address dynamic optimization problems. They either assume that changes in the environment are known a priori, especially for some benchmark problems, or detect these changes. On the other hand, detecting the points in time where a change occurs in the landscape is a critical issue. In this paper, we investigate the performance evaluation of various sensor-based detection schemes on the moving peaks benchmark and the dynamic knapsack problem. Our empirical study validates the performance of the sensor-based detection schemes considered, by using the average rate of correctly identified changes and number of sensors invoked to detect a change. We also propose a new mechanism to evaluate the capability of the detection schemes for determining severity of changes. Additionally, a novel hybrid approach is proposed by integrating the change detection schemes with evolutionary dynamic optimization algorithms in order to set algorithm-specific parameters dynamically. The experimental evaluation validates that our extensions outperform the reference algorithms for various characteristics of dynamism.