数据分布诊断对排队模型优化影响——基于兰州市某油库排队问题数据

Data Distribution Diagnosis’ Influence on Queuing Model Optimization——Based on Queuing Data at One Depot in Lanzhou City

对甘肃省兰州市某油库真实排队数据进行分布诊断,发现用M/G/C/∞排队模型计算出来的平均队长和平均等待时间比M/M/C/∞排队模型减少了一半,应根据数据分布诊断结果科学选择排队模型。

Queuing problem is serous at a oil storage in Lanzhou city Gansu province. This paper made diagnosis on raw queuing data, find out that： compared to exponential distribution service time queuing model, if choose the common service time queuing model, average number of customers in the queue and average time customer spends in the queue can be reduced by half.