ADVANCED MANUFACTURING NOW Manuel Terranova Modern Manufacturing Processes, Solutions & Strategies
TIME TO ‘ABSTRACT’ DATA SETS FROM ALL OVER INTO ‘DATA LAKE’
I
t’s no secret that data being generated by sensors embedded in equipment has opened up a
new world of possibilities in manu- facturing. Advancements in storage, bandwidth, processing power and algorithms enable us to mine data quicker than ever. With some Global 1000 manufacturers already realizing $1 billion of revenue from telemetry, you might think all of the pieces are in place to carry out the digital indus- trial revolution. However, the reality is that telem- etry alone can only deliver a fraction of the possible predictive mainte- nance capability and breakthrough innovation. To perform meaningful advanced analytics, manufacturing engineers need a complete set of data, and sensors only provide one piece of the puzzle.
Completing a Puzzle Telemetry data is undeniably an
essential ingredient, but to truly mine its value it needs to be com- pared to other equally important very large unstructured data sets— namely the original equipment’s geometry, schematics and simula- tions. If these three data sets could be viewed together, the telemetry data generated on the test bench would potentially validate original designs or illustrate their flaws. Likewise, simulations of how parts behave under certain loads or condi- tions also inform the manufactur- ing process when examined against telemetry. Over time, engineers can continually refine designs based on
telemetry readings from the test lab and the field. Unfortunately, today’s IT in-
frastructure prevents the people who need these files from access- ing them. From the time these files get created, engineers have a hard time getting them out of the stor- age buckets and silos in which they were initially placed. Moreover, the IT department eventually moves them around the enterprise as part of periodic hardware and software refresh initiatives. Thus, a simula- tion or drawing created in these early stages of the product lifecycle
on a server in Italy and a simula- tion housed in the cloud in a single view? The truth is that it would take several days at best, but more likely many weeks, to get these massive files together in front of the people who need to make sense of them.
Unifying Data Access Large manufacturers cannot
control the fact that data sets reside in disparate locations. However, if they want to make aggregation and access a priority, there are ways to “abstract” data into a single “engi- neering dataplane” or “data lake”
FROM THE TIME DATA FILES ARE CREATED, ENGINEERS HAVE A HARD TIME GETTING THEM OUT OF THE STORAGE BUCKETS AND SILOS IN WHICH THEY WERE FIRST PLACED.
can be effectively lost by the time equipment is in the test lab and out in the field—five years could elapse between the completion of an initial drawing and the first prototype’s arrival on the test bench. By that time, engineers have to go to great lengths to track mission-critical files. Today, software-defined solutions and cloud offerings have made the storage part of this challenge much simpler and cheaper to address. However, IT and engineering leaders haven’t made access and aggrega- tion of unstructured data housed on storage solutions a priority. How are engineering teams supposed to assess telemetry data coming off the machine, a geometry drawing lying
from which engineers could access groups of files as if they were located on a single hard drive. This approach streamlines the process of scouring the enterprise for every mission- critical unstructured data set, but puts manufacturers in a better place in the long run. Once the founda- tion is put in place to abstract data, engineers will have the full picture needed to fine-tune design and man- ufacturing processes throughout the product lifecycle. Sensor readings from the test bench and throughout the equipment’s time in the field can be aggregated with design and analysis to provide a complete set of data to bring forth the real promise of the Industrial Internet.
43
President and CEO Peaxy Inc.
Spring 2016
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70
