May 4, 2016 at 4:29 pm #2681
I worked with a group throughout this past semester to model and analyze the process of using the ATM in Union South. We tracked the arrival time, wait time, and time in the system during three afternoons. Each day we had about 50 customers in order to make sure we had enough data for it to be accurate.
We then wanted to find out if this system was the best possible. To do this we came up with two alternate models. Alternate one consisted of one queue and two machines, alternate two had two queues associated with the two machines. We looked at the utilization and average wait times of the original and two alternates.
We found that the average utilization of the original system was around 50% and for the two alternate models the machines were each at around 25% utilization. While the wait times were decreased by about 30 seconds, on average, we concluded that the implementation costs of adding another ATM machine were not worth it.
Do others believe that implementation costs should have as much importance as convenience? I’m curious to know what other people value based on their standpoint on the situation (customer vs. manager).May 6, 2016 at 1:53 pm #2688
This was a very interesting project, and cool to investigate the queue times for an ATM on campus, which many students use. I think what was different about this project from what the rest of the class is you could have a high utilization without overworking the system, as long as queue times were low. For example, other groups did projects in the food industry where the “machines” were workers. For these cases you don’t want to have a high utilization because the workers need time off for lunch and breaks. However with an ATM it doesn’t ever need break time so you can have a higher utilization than you would expect for actual employees. I think since the queue times for this ATM were pretty low, the cost of adding an additional machine would not be worth the small benefit they would provide. Perhaps it would be cost effective if it were highly used all the time.April 16, 2018 at 5:06 pm #3462
It will also be interesting to look at ATM usage as pertinent to cash usage vs card usage in that area. With the increasing using of non paper money, I wonder how useful ATMs will be in the future.April 16, 2018 at 6:05 pm #3470
To answer whether implementation costs outweighs usage is largely contingent on utilization rates. In the case that is outlined above, I think that it is very important to take into consideration these kinds of hard to quantify measures. You could look at a quantifiable statistic like balks or you could attempt use utility to determine if the reduced wait time makes a large difference.
You must be logged in to reply to this topic.