Improving operational safety and efficiency of variable speed limit (VSL) systems are the two core control objectives at work zone areas. In response to such need, this study presents a proactive VSL control model for freeway work zone operations. The proposed model uses an embedded macroscopic traffic flow model to predict the traffic state evolutions over the projected time horizon and to determine the optimal speed limits. In addition, Kalman filter is adopted to correct the prediction inaccuracy in a timely manner. To improve the safety of operations, this study proposes a new control objective function to smooth speed transition along the target freeway sketch by minimizing the difference between actual speeds and ideal speeds. Also, the smoothness of speed transition can help prevent the formation of shockwave and consequently enhance system’s operational efficiency. The authors’ numerical experiment with a calibrated Verkehr In Städten–SIMulationsmodell (VISSIM) simulator reveals that the proposed VSL system can significantly reduce the speed variance among different freeway subsegments. The evaluation of several measures of effectiveness (MOEs) also shows the promising results of the VSL system on the improvement of freeway operational efficiency.