How to Cut Maintenance Costs with Big Data
With bogie maintenance now accounting for around 12% of the total vehicle costs it is not surprising that so many are looking for ways to ensure work is conducted at just the right time. This is now possible thanks to condition based maintenance and predictive analytics.
Gerald Schinagl: “Massive Cost Advantage”
In a recent interview Gerald Schinagl - Digital Innovation Manager at ÖBB in Austria - talks about the impact of condition based maintenance for the railway industry:
“Knowing and predicting the demand for maintenance enables us to completely rethink how rail operations are supplied and what amount of rolling stock is needed for that. It enables much more planned operations.”
Today advanced sensor technology plays a crucial role in this strategic shift, especially interlinked with Big Data driven analytical algorithms.
From Data to Insights
Data integration is about managing complexity, streamlining connections and making it easy to deliver information. Today it is already possible to turn data into information, information into insight and insight into very real cost savings. Big Data will change the rail industry forever, in the same way is did aviation and road transport. The revolution has already begun.
Big Data in four V’s
The information and value that we get from Big Data is hidden within four V’s:
- Volume: the scale or amount of data that is being produced
- Variety: the diversity or changing nature of that data
- Velocity: the vast speed at which this data is arriving
- Veracity: the accuracy or quality of the data
In the past the time taken to analyse data resulted in costs that were too high to be of value. Now however, the rugged sensor technology is a lower cost and more qualitative. It now offers an unprecedented ability to collect and analyse multiple data sets and compare entire train fleets, in real time if required. In that way we can bring together information from real time dynamic data sets (unparalleled in size) as well as historical information to reveal never before seen results.
As the new technology can be applied to both new trains and legacy rolling stock, it enables the combination of data sets for even greater insight. These intuitive systems provide accurate reports on the performance of individual components such as bearings, gearboxes, wheels and even the complete bogie structure. Applying location and temporal intelligence to a static dataset makes it spatially comprehensive and offers new levels of insight around that dataset. It helps engineers to understand and comprehend wear and tear on a whole new level, giving them the tools to predict wear and failure, leading ultimately to cost savings.