How to predict the next date of a customer visit?

Question

Analytics

There is an array of distances between the dates of the visit by the client, of the form, for example:
came after 5 days, then after 5, after 10, 7, 20, 30, etc.
Based on this series, it is necessary to predict when the client can come next time (it is necessary to track when the client did not come at the estimated time - that is, there is a chance that he will become a lost client and he needs to send an SMS reminder.) We are talking about a beauty salon where Clients visit fairly regularly.
Tried Median, arithmetic mean.
Are there other methods or formulas? I would like the technique to ignore cases when the client went on vacation or something else.
Upd: I would like a simpler solution, without neural networks.

Answer 1

I'm helping a friend automate a beauty salon. So, for each service, he set the time after which it is necessary to call the client and invite him to repeat the procedure.
Simple and tasteful.

Answer 2

• Overview of Time Series Forecasting Models
  The model will be able to predict the probability that the client will return after a long absence (left on vacation) only if this client has already returned after a long absence or if this client is assigned to a group of clients who were previously characterized by such behavior.