In recent years, crowd simulation has received a surge in attention. One of the reasons for investigating crowd flow is to optimize evacuation procedures from densely occupied buildings and urban centers in case of an unpredicted event. Such evacuation planning can save many lives at sudden catastrophic events like tsunami. In this paper, the effectiveness of phased evacuation strategies is examined in a large-scale exhibition hall facility, under the assumption of both full and partial compliance of evacuees with the evacuation instructions. It is clearly demonstrated by numerical analyses that phased evacuation has positive effects on the mitigation of high-density pedestrian flows and heavily congested state in comparison with traditional simultaneous evacuation. Our approach has potential for contribution toward investigating disaster risk reduction strategies in complex urban space.

Exploiting the effective utilization of the small-cell size will become an enabler to meet the demands posed by data explosion and energy crises for the ongoing 5G cellular networks. In this paper, we propose selfoptimizing based energy-efficient solutions for densely deployed and on/off-aided femtocell networks. The proposed scheme aims at achieving high SINR with a tradeoff between energy efficiency (EE) and spectrum efficiency (SE) by adaptively coordinating the downlink transmit power based on a Neuron control process. The optimal coverage of femtocells can be obtained from flexible cell sizing to cater for the instantaneous traffic requirements. Accordingly, unnecessary femtocells can be switched into sleep mode for more energy savings. To cope with the topology uncertainty and deliver the most appropriate access operation, the femtocells are dynamically configured as clusters in which a modified K-means algorithm is performed. The utilities of all the action candidates are jointly estimated by which the coverage extension and on/off operations can be fairly performed with a maximization of the cell-level and cluster-level EE. The numerical results show that the proposed self-optimized solutions can effectively reduce the downlink energy impact with less active ratio and optimal coverage of the femtocells meanwhile satisfying the throughput requirements compared to the conventional configurations with regular and predefined patterns.