You are not crazy just because you talk to your fridge

Aug 4, 2018 10:25:00 PM / by Jag Padala, Chief Technology Officer

Innovation comes in various flavors. Revolutionary, evolutionary and incremental. The connected web has been a revolution that is on par with any other innovation in the history of mankind. When revolutionary change happens it can spur a whole host of evolutionary and incremental changes.

The Internet of Things (IoT) and Health Care

The world of IOT is happening at a vary rapid rate. The need to monitor systems and devices has always been critical and a needed physical presence. This is even more applicable in a health-related setting.

Monitoring Devices and Tracking Anomalies and Exceptions

Monitoring devices and tracking anomalies and exceptions is just the beginning.  The resolution of the exceptions may need a number of disparate stake holders to be involved.

Social Collaboration

The evolution of social collaboration is paving the way for the possibility of the next generation of exception handling mechanisms. The simple page has been replaced with group text messages. With the evolution of social networking tools this is rapidly moving into the next phase of possibilities.

Example: Hospital Refrigerator Banks

A mid-size hospital typically hosts a bank of refrigerators to store a variety of products. The materials are extremely temperature sensitive and hospitals have regulatory requirements to have tracking and exception handling systems in place.

Once the basic systems are in place there are two avenues for revolutionary change.

1. Prediction of Refrigerator Failures

The first is prediction of failures and is already in place in most sophisticated hospital systems. The rate of failure can be predicted based on various characteristics such as the model, age, load and this can provide the facilities department with a reasonable estimate on the time and  effort that would be needed for the maintenance activities.

2. Talking Directly to the Refrigerator to Obtain Critical Information

The second however is more interesting and the focus of this article.

When a failure message is sent to the stake holders each of them have various data points that they would like to get access to.

  • The facilities manager would want to know if a technician has acknowledged the failure.
  • The technician would probably like to know how the refrigerator has been behaving in the time leading to the failure.
  • The technician would also like to know if replacement parts are needed and if the are onsite. If the parts are not onsite does the supplier have them in stock.

Normally these activities would involve a physical visit to the premises or at the minimum direct access to the system to get the detailed information.

The evolution of chat is providing a great avenue to get information by directly talking to the system. This in fact has huge life style repercussions.  Having the ability to execute these kind of activities from a simple smart phone gives the stake holders a great flexibility in their life style. They could be at a kid’s soccer game and still attend to an emergency with a few clicks on their smart phones.

Chats in Two Flavors

The chats are of two flavors.

1. Key Word Commands

Key word commands are the first in the line. The specificity of key word commands makes it unambiguous for the system to respond.

The TAGNOS temperature monitoring system for example provides a set of key words such as @showhistory @showchart @showfailures @includesupplier. These keywords are included when each chat room is initiated at the beginning of the incident. Users can type in the key works to talk to the system directly to find out the history of temperatures for the refrigerator, get a chart of the readings, get a quick chart of the readings, show failures that have happened on the device in the past and include other stake holders in the conversation. More commands like @checkpart xyz can also be provided.

Using these commands technicians can diagnose and troubleshoot issues remotely. All the conversations pertaining to the incident are automatically archived for future reference.

>> See Asset, Supply & Temperature Tracking

2. Natural Language Based

The other flavor of interaction is natural language based. Using templates that interpret the natural languages the sentence is de-constructed into commands.

Using just structured commands is itself proving to be a great winner in terms of response speed, closure intervals and convenience.

>> See How A.I./Machine Learning Helps You

Talking to Hospital Refrigerators is Far From Crazy!

TAGNOS is a company dedicated to improving patient experience by addressing operational issues at hospital and thereby providing better patient care. And, that includes speaking directly with refrigerators.

 

Improve OR Turnover Times with RFID Download the Presentation