How wearables and footballs helped me “outthink limits”
Day 1 of IBM InterConnect on Sunday saw IBMers and IBM Business Partners from around the globe get together for the Business Partner Summit.
The opening general session held over breakfast at the MGM Grand really established the ‘outthink limits’ theme for the five day event. While the idea of digital disruption isn’t new to any of us, it’s sobering to consider the many shapes and forms disruption can take, and the consequences of looking the other way. The story of the Swiss Watch Industry is a case in point – where continuing to dismiss the potential impact of smart watches and wearable devices was met with declining sales and market share.
I found this example a poignant reminder that a cultural and personal willingness to imagine, create and embrace change is just as essential to ‘outthinking limits’ as technical innovation.
IBM Watson IoT Platform:
Cloud-Based Device Integration for Business Applications (Peter Crocker, IBM UK)
I was expecting IoT (Internet of Things) to be one of the dominant themes of the event, and I got stuck straight into it in the very first presentation of the day.
What I found particularly interesting was the low cost of sensors and the amount of information that can be provided. There are some challenges with security, however IBM is investing in research on the use technologies available with digital currencies, such as bitcoin, to build this integrity (see blockchain and IoT).
Another interesting example Peter provided was around the use of a simple Bluetooth speaker with a microphone as an IoT device. You can connect the device to the Internet and talk to it. This will call the IBM Watson API for speech cognition in order to interpret the command, execute the operation and respond with another call to the text to speech API.
IBM have also acquired The Weather Company which allows yet another dimension of data to be added. Can you imagine reviewing the maintenance of assets based on the conditions under which they operate? This adds a whole new perspective on understanding the performance of an asset. IBM Weather Insights even includes data collected from airplanes!
Peter went on to perform a demonstration using a Texas Instruments (TI) SensorTag, which is essentially the size of a USB drive and includes around 14 sensors. Some of the measurements include humidity, acceleration for the each of 3 axis (x, y & z), temperature, pressure, etc. Peter had the sensor stitched into an american football and proceeded to throw it around the audience and we watched in realtime as the sensor via Bluetooth updated the Cloud web application with results.
IBM Predictive Maintenance and Quality:
Improve Asset Productivity and Product Quality (Ishan Sehgal, Program Director PdM)
This was a very interesting presentation and is certainly a key field of interest for Clarita’s product teams. Predictive Maintenance is driven from 5 core principles:
- Collect data from multiple sources (operations, asset attributes (supplier, manufacturer, batch,…), weather, etc)
- Correlate data (summation and pruning of data)
- Optimisation through the development of models and building asset profiles
- Scoring to find patterns / modelling actuals which provides decision options or recommendations
- Action to drive business outcomes (safety, reliability, performance, capex, opex)
Ishan discussed the approach used to develop the product which identified different personas (manager, operator, etc) and how they viewed the problem statement and measures of success.
Automated analytics is simply the orchestration of tools such as SPSS execution of a model on assets wrapped in python scripts. The level of automation depends on the domain knowledge available to know what model to apply to solve the problem. This knowledge is available as “tribal knowledge” or the business rules that have been built up in an organisation over time. For example, after three stops, an inspection and clean is required.
Ishan spoke extensively on the importance of master data management and how the processes to update data are critical to the success of IoT and Predictive Maintenance (PdM).
Manufacturing Maximo User Group:
Ford America’s Experience with Maximo Scheduler Best Practices and Optimisation (Elaine Staton & David Borland)
This joint presentation by IBM and Ford addressed the topics of Capacity Planning (resource leveling) and Labour Assignment (changing from a requirement of a Craft to an actual Crew or Person).
It also highlighted some key data requirements to ensure the Scheduler Optimisation engine (ODME) can run successfully. Some of the required data includes:
- Labour rates – to assist with the scoring (less expensive labour ranks higher);
- Calendar and Shift records as well as any shift breaks and far enough into the future that is required by the model; and
- Craft, Crew and Labour records so they can be used on Job Plans or Work Planning.
Key takeaways from the session included:
- Plan the resources at the level they are managed. So if they are managed as Crews, perform the planning at the Crew level.
- The duration of a work order affects scheduling and cost. Duration is the time taken to complete a work order. So for example two plumbers each work two hours simultaneously for a total of four hours of labour, but the duration would be only two hours.
- Use Service Addresses to calculate travel times and also setup Crew start and end locations so this can be included in the calculations.
- The Scheduler product is moving away from requiring a Java client from Maximo version 184.108.40.206 and 220.127.116.11.
Business Partner Watson IoT Reception
While the sessions are always packed with great content at events like this, there can be just as much value to be found in getting together socially with others interested in Maximo & Tririga. Wrapping up the day at the Watson IoT Reception provided just such an opportunity and I had the pleasure of meeting a number of like-minded IBMers and partners. Thanks for the catch-up Paul Lee, Gareth McGlone, Phil Gruber, Peter, Johannes and Louis Stoop.