Bartender smart spout IoT device

Initially, a startup company came to EnCata with a very crude POP (proof-of-principle) prototype of the smart spout device built on Arduino, seeking design and engineering help in developing their product. The initial product idea was to provide private customers with a device that can tell them the amount of alcohol consumed and a tool for social interactions.

EnCata started this project from the technical feasibility study, choosing the technology for spirits and liquor metering. It soon turned out that new technology had to be developed from scratch. Thus far, a new concept was developed and EnCata’s engineers started to develop the alcohol metering module, which had to precisely measure wines, spirits, and viscous liquors containing sugars.

Upon completion of the R&D, a new industrial design was developed around the internal modules of the smart spout. The PCB received IoT functionality by adding BLE (Bluetooth low energy) and NFC:

• the BLE provided connection of the device to the smartphone, that acted as a gateway and the mean of connecting to the cloud;
• the NFC tag administered the spirit/wine bottle selection 

After spinning out the PCB, embedded software was developed and deployed to the hardware. And the firmware team started work together with the cloud developers to bring up the API. From this point, the hardware and software engineers started to work in parallel: 

1. The software team started developing mobile and web cloud applications, that provide the user all the statistics of the alcohol.

2. The hardware team continued to refine firmware and the mechanical engineers commenced the DFM, as part of the engineering validation testing (EVT).

It is estimated that bars (and, in turn, distributors) are losing up to 50% of their profits due to: mispouring by the bar staff, conveying unrecorded sales, low thoroughoutput during the peak time, running out of stock.

Upon the EVT completion, the device quickly passed the DVT, while the client’s team worked on Kickstarted production. The “Smart Pour” device featured at the Kickstarted as a B2C consumer product and sought a very modest reception.

In response, EnCata suggested pivoting the product and pursue the B2B market. A new business model was suggested, which required a new type of device to be constructed. The new [group of] devices relied on previously developed alcohol flow metering technology, building a whole ecosystem around the smart spout, converting it into the point-of-sale (POS) accepting cashless transactions.

The new use-cases aimed to incorporate the precision spirit metering technology, NFC-enabled payments + self-service, and enable supply chain control (from the very bottling station at a distillery) for every bottle.

For that, some of the whole industrial design of the device had to be revamped. New features incorporated in the spout design (snaps, screws, and bottle fittings) required high precision and superior surface finishes, in order to suit the engineering and design tests. Thus, all the mechanical parts were developed for injection molding, as early as for the EVT. Several molds were developed (in-house) and a few hundred parts were manufactured using our semi-automatic injection-molding machine. 

On the software side, several APIs for payment systems (such as STRIPE), ERP, and POS (client application) talking to the server application. In this architecture, each device (client) talks to the web-server (located on a special cloud machine) via mobile or web-application.

The server runs two separate application: 

1. A cross-platform application containing business-logic built with Ionic Type Angular technology (supports both mobile devices and browsers realized with common HTML, CSS, and TypeScript). The Ionic Type Angular is built on top of the RxJS library.

2. A non-relational database (Java 11 c with multimodular approach including Gradle, Spring Boot, Spring Data, MongoDB, TomCat, Docker, Nginx).

The remote control framework is configured with JMX. And JMS compliant API to ‘listen’ and processing of client messages. The software additionally uses MQTT broker to convey messages between the client and server application. 

Thus server can simultaneously talk to many clients (devices) ensuring high reliability and performance.

There are 62,600 bars, taverns, night clubs in the USA alone. They generated $28 billion in sales in 2018. 35% of total sales is spirits and liquors ($9.8 bn).