There is no denying that there is an all important human aspect to the process of ordering coffee. Whether it is the person who rings up your order at the register or the one who calls your name with the drink. That being said, there are areas where human interaction is not needed and a computer would be better suited to do the job. The first wave was provided by companies like Folgers giving the United States instant coffee and the like. Then Coffee was about convenience. Starbucks was the first to bring high quality Italian coffee to the United States and to present that style of ordering.

Main Problems:

• What Starbucks also brought in were unseen amounts of customization to their drinks which is part of their fame. Indeed Starbucks boasts 87,000 permutations of their drinks.

• Another part of the Starbucks fame has been the inability for Starbucks Employees to properly write the names of customers on the cups. Evidenced by our experience at the University of Rochester as both students ordering drinks as baristas making the drinks we see that there is an overall need to improve the ordering process.

• Language processing barrier

We have divided the improvements into a few categories. The first is to make coffee more accessible. There is a language to order Starbucks drinks all on its own including important syntax and semantics. The phrase “Grande Hot Mocha” is a well formed drink but so is “Grande Brown-Sugar-Oat milk-Shaken Espresso.” In addition to the details required to order a drink which can be confusing enough, there can also be language barriers introduced that complicate the process which can make it more difficult for the baristas to understand what is being said and more difficult for the person ordering to get what they want.

By using an ordering system that makes use of such a system that accelerates the pace of ordering by removing the overhead needed to train baristas on all of the many codes that need to be understood to make the drinks.

A machine would be able to type : Latte with soy milk and two pumps of vanilla faster than a person could write: L + S + 2V in sharpie and ring the drink up on the register.

How so?

It does so by providing a natural language processing system that will uses existing ordering methods: such as a drive-through or voice to text. This stands as a way to simplify the ordering process and enable employees to be allocated to other areas such as the more important making of the drinks themselves because it provides live verbally-affirmed feedback.