Invent

No Data, No Digital

The joint efforts of IT and production professionals have proven that it takes time to become an industry 4.0, with much experimentation, development and learning along the way.

Nov 2017

Forget for a moment about slim screens, augmented reality and hi-tech accessories. An industrial environment is highly complex. You need more than just a futuristic set-up and the very latest, state-of-the-art machinery to turn it around. What you need first of all is the raw material required for digitalization and to make interconnection possible: data.

This raw material, like any other, must be treated and processed with skill, and represents the second essential aspect for digital industrial transformation: revolutionization of how we work. The situation at the facilities of Avio Aero is, therefore, one that no-one would have believed possible just a decade ago: more computer technicians, programmers and digital specialists (like the specially formed team of the Digital League) are turning to the factory. Turning, that is, to the machines that make the parts of engines, to their mechanical and electronic operation, to the work cycle of aeronautical products, and to the typical indexes of Lean manufacturing. Similarly, Global Supply Chain (GSC) engineers and operators have begun to handle advanced information technology, and their experience will, over time, revolutionize the way they interpret their work. 

Anyone following the evolution of what is known at GE as brilliant manufacturing will certainly understand what we are talking about. At the same time, the Brilliant Factory can be seen as an “industry series” of new scenarios, discoveries, developments and introductions of new tools over the months. These are not an end in themselves, and not to be taken purely at face value; they are very real tools that operators, engineers and technicians are familiar with and know how to use, and which they know they can trust.

The facility of Rivalta di Torino covers over 130 thousand square meters and employs about 1500 workers in eight areas or fields of production. Five of these areas use a digital operational tool called EVO (Electronic Vouchering) known on site as on-time vouchering. Within a few months, each of these eight areas will have this screen in its own production cell, including a touch screen which each operator of the cell has begun to use. The operator signs in at the screen with his badge and can view and update (add to) the WIP (work in progress) data for each part of the engine being made, entering in the information system confirmation of the completion of any given operation.

WIP is the sum of the parts that are brought to the factory to be made into finished products. “We organize operational training for everyone who works directly on the machinery and more in-depth training for production technicians and engineers,” said Davide Mazzucco, IT Project Manager. “Operators did not use to handle this type of information; rather, the production supervisors or engineers were the ones who did that. Currently, each step of the cycle has its own bar code which physically refers to the part, or piece, and is digitally ‘tracked’ thanks to EVO.” 

Anyone at Rivalta now who works on a part of an aircraft engine can refer to the number and type of operations previously carried out on the part, be aware of the equipment used for machining, and enter information on the work done personally. “This application has found a highly suitable and diverse production environment in which to be tested and used,” explained Antigone Barbera, GSC Leader of the Digital League. “The next step will be to adopt it at the other facilities where operators do not yet use touch screens for entering data. We would then like to use this electronic data system to replace our current stamping system for paper cycles with what we call Digital Stamping”.

Everything is done in real time and, above all, shared and made available on a network. This is because all the information, meaning every operation, is converted into data. Also, the progress of each and every part - and consider that several hundred parts are produced at Rivalta alone every month - generates data in turn. 

This is why not only those who handle measurements and production indexes, but also those who physically work with parts under production, have begun to handle data. So much so that, at Avio Aero and GE Aviation, the term “Data Lake” is typically referred to by those who are responsible for production as well as by digital engineers and IT specialists.

“A few years ago, we first called all data generated and collected in the context of manufacturing our Data Pond. The volume back then was of course much smaller, but now there almost literally is a lake, our own Data Lake,” said Alberto Riccabone, IT leader at Pomigliano d’Arco, who for some time now has worked increasingly among the lines and cells at the product centres in Campania. “The Data Lake contains millions of pieces of information relating to our GSC, a constant flow of which comes from every department of the company. For a while now, the site in Pomigliano has made its operators familiar with the concept of traceability and an app has finally been developed and adopted in some areas for managing production priorities with dashboards. These are control panels on large screens.”

“A few years ago, we first called all the manufacturing data generated and collected in the our Data Pond. The volume back then was of course much smaller, but now there almost literally is a lake, our own Data Lake.”

A “business intelligence” program, also used at GE Aviation (it is, therefore, standardized and shared worldwide among the businesses of the group) and which retrieves information from the Data Lake, is used by the IT teams, together with those of Lean Manufacturing and Digital League, to create a variety of interfaces and applications (depending on the nature of the data retrieved and handled) for the same GSC teams.

We are talking about WINS (What Is Next, Scheduler), the evolution at Pomigliano of a former online system called Dyoper: “it was essentially an Excel file which, when rewritten with this visual reporting program, looked like it had been treated with steroids!” said Alberto. WINS is now used at some cells at the Combustors Centre of Excellence and the CRO (Component Repair and Overhaul) workshop in Naples for managing priorities and programming the job queue for parts that follow a production process (Combustors), or modules and components that require repair or overhaul (CRO).  WINS sets the priorities in relation to the targets, helping the operator organize the work according to the timeframes and, therefore, the commitments agreed with the customer.

“The factory workers at Pomigliano would typically rely on verbal reports or track the parts with tags and paper records,” explained Gianni Arbia, GSC Materials Leader at Pomigliano. “The system was time-consuming and the workers found it frustrating. It also increased the risk of losing information.” For the local team, this is a radical change which will extend to all the cells of the site by mid 2018. It could, therefore, also be adopted at the other facilities.

The choice of CRO as the pilot area for this tool was by no means random: “Our workshop has particular work schedules and processes that do not follow traditional production logics,” said Sabatino Covone, CRO Operations Leader. “Perhaps with us being the first to test and adopt these innovations, not only does our team have a chance to evolve, like with the intelligent eyewear, but also the developers of these tools are able to test adaptability to very different contexts and processes.” 

Old approaches and methods interpreted and organized afresh by generating, processing and sharing data. This is the case of another ambitious project in Brindisi, where automated and interconnected lines with new generation software run alongside machining equipment (like lathes and milling or grinding machines) that, despite being older, “guarantees exceptional precision and finishing, still works perfectly and is also sufficiently flexible to be made digital,” according to Antonio Reale, Lean Manufacturing Engineer at the site in Salento.

Antonio, one of many young production engineers emerging in the field of artificial intelligence for machinery, said: “I won’t never be a programmer, but I am enjoying working with the IT team and the results of the first experiments have been particularly rewarding. Every day there’s room for new cue.”

The project in Brindisi is quite intriguing: with some imagination, it could be called a simultaneous translator for machines that allows them to speak the language of data. It should first be pointed out that, at factories over a century old like those of Avio Aero, there are at least three types of machinery: new generation and ultra-modern (interconnected, with digital setup as mentioned before), numerical control (the most common, with electronic control interface) and analog (with manual and more basic controls). The latter account for about 10% of all the machinery at Brindisi and a highly successful mission is underway to save what are perfectly functioning machines that would otherwise be brought out of service simply because of their inability to generate and send data, saving several hundreds of thousands of Euros in the process.  

"We can bring older machines up to the level of more modern ones with gateways, sensors and network connections. This is in the interest of interconnection."

The mission, which for a few months now has been carried out in the new Repair area at the site in Brindisi, and specifically in a cell for older lathes and milling machines, effectively involves connecting the machines to an online network for the integration and sharing of the digital language. Angelo Spalluto, the IT leader at the site in Puglia, said: “Our experience so far with the various types of machine and the ultra-modern ones for the LEAP, and our vision of an intelligent and integrated shop floor, have motivated us to develop solutions also for special applications like older machines: because we can bring those up to the level of more modern ones with gateways (in the context of information technology, these are routers that manage input and output data, ed.), sensors and network connections. This is in the interest of interconnection.” The results of the first tests have been encouraging: “we put a box on an old-school machine, and connected the sensors that send electrical signals for conversion into data by the gateway,” explained Antonio, now well versed in what is literally a complex process.

Engineers who become computer experts; nerds specialized in the binary system who become proficient in mechanics; operators who not only handle heavy equipment but also use touch-sensitive screens; and ‘70s machines that “speak digital”: this is what transformation is all about