Good morning,
I'm trying to assess whether Alteryx is the right solution to build a queueing model (customer arrival rate, number of servers, wait times, utilization rates, etc.). A colleague mentioned that it might be a better option than, say, Crystal Ball, but was unable to elaborate further.
Would anyone be able to weigh in on how one might think about building a queueing model / simulation in Alteryx? Thanks!
Allen
Solved! Go to Solution.
Hey @allenxu,
IF you're looking for a process simulator (where you shove in different load factors and try to see how the process performs - like the simulators of a McDonalds layout to see which creates the fewest backlogs), I'd probably look to a dedicated solution that can do this dynamically. I've been on projects before where a package called VenSim was used (years ago - http://vensim.com/), and I've seen this done in Signavio, but not sure of the market and players in process sim right now.
Alteryx would probably be able to do a static-state version of this - but it's not engineered to do the things that the large scale process simulators do naturally (like different load distributions; mean-time to failure on each step; max queue wait time before abandonment etc). These capabilities are important for large-scale industrial processes, because you end up with non-linear behaviours that can only be seen in dynamic simulation.
To be clear - not trashing Alteryx, it's an incredible data prep tool - but it's not built to be a process simulator so you'd miss many of the core capabilities you'd need.
Cheers
Sean
I would echo the advice from Sean Adams. Another tool you could investigate is JaamSim at http://jaamsim.com