This study uses simulation-optimization and Reinforcement Learning (RL) to analyze the routing behavior of delivery vehicles (DVs). By conceptualizing the system as a stochastic k-armed bandit problem, the RL model helps DVs modify their routes based on delivery strategy selection. The experiments conducted on a simulated network with realistic traffic conditions show that employing an RL-based decision support system for en-route decision-making enhances the overall efficiency of the transport network.
内容由零声教学AI助手提供，问题来源于学员提问
![local s = ""
local sstr = string.gmatch(s,'href%s*=%s*"([^"]+)"')
print(sstr()) 增加一个功能，左右两边可以是双引号或者单引号](https://linuxcpp.0voice.com/zb_users/upload/2023/05/202305162224299866311.png)
![local s = ""
local sstr = string.gmatch(s, 'href%s*=%s*["\']([^"\']+)[\'"]')
print(sstr()) 增加一个功能，开头可以是href、src、action](https://linuxcpp.0voice.com/zb_users/upload/2023/05/202305162217558813401.png)
