Why a production process needs information technology
The age of industrial security - Information technology has changed more radically over the past 20 years than practically any other field has.
Information technology has changed more radically over the past 20 years than practically any other field has. Particularly in connection with the Internet and mobile communication technologies, IT has progressed at breakneck speed to become firmly entrenched in every aspect of the modern factory. It is only with the widespread application of these technologies that it has been possible to network production facilities, plants, and machines across different locations, supply chains, and national boundaries. This development has brought a boost to productivity comparable only to the advent of automation technology.
Henry Ford famously said: "People can have the Model T in any color—so long as it's black." In Germany, a potential buyer looking to purchase a small two-seater car back in 1924 had the choice of just two producers and two colors: light green or yellow.
Buying a car in this day and age means choosing between a seemingly infinite number of options: style, passenger capacity, colors and finishes, engine sizes, interior trim, and equipment variants. The variety is so enormous that statistically speaking, a year will pass before an automaker will manufacture precisely the same car twice. Controlling this enormous model diversity requires a high degree of automation in production and an equally high degree of networking on every on level of the automation pyramid.
This starts on the business administration level with purchase order processing and the rough planning of material input and production, through the control and automation level with detailed production planning, operation and monitoring, quality management, formulation management, and so on, as far as the field level that forms the interface to the technical process itself with its execution, control, and regulating functions. Only by ensuring the smooth interaction of all these aspects can the production process effectively manage each individual customer order and comply with every specific request.
To ensure trouble-free interlinking of the countless different process steps, a large number of data records must be exchanged between the various stations of a process chain. To stay with the example of car production, this starts by assigning each new car’s floor assembly a unique number, comparable to the engine or chassis serial number used later for vehicle registration. This identifier is applied during the assembly phase and stays with the vehicle until it rolls off the production line.
At every station, from the body in white, through the paint shop to the trim shop, during interim storage at the various stages of variant production or during final assembly and inspection, the control system and line operatives know exactly which vehicle is being produced when and in which location. This is not as straightforward as it sounds, as every scanning process has to take place at every stage of production and in all types of adverse conditions, even at temperatures in excess of 392 F (200 C) or continuous movement of the assembly line.
The same underlying principle applies across the board: The more individual the vehicle’s equipment features, the more information the car manufacturer has to collate, process, and evaluate during production and send to the different controls, drives, and sensors to ensure that the right car with the right paintwork finish and equipment package rolls off the line at the end. A robot manufacturer has described how its robots are required to weld 6,000 different seams with an accuracy of 0.2 millimeter on a single car body, noting 150 different model variants. Data picked up by the sensors relating to welding point accuracy, energy consumption, torque levels, rotary angles, and so on keeps operators informed at all times about the status of production and manufactured products. Considering that one new car rolls off the line every minute—adding up to 1,400 times every single day—this information is vital.
Controls and networking
When the first programmable logic controllers were used in production 50 years ago, they had one objective: to provide an electronic and freely programmable alternative to the failure-prone relay. Up until this point, relays had controlled all the automatic processes of a production line, in a similar way to their function in the Konrad Zuse’s first binary computer, the Z1. What followed were durable, compact units that could be distributed around the plant and linked by means of networks.
Initially, these were proprietary networks that afforded little compatibility to the devices supplied by other manufacturers. But still, the use of distributed intelligence represented an enormous step forward in the development of machine and plant engineering. The precision, speed, and durability with which processes could now be controlled were also impressive. Since this time, automation systems have controlled not only robots, machine tools, agitators and filling stations, rotary kilns, sewage pumps, rolling stand drives, paper machines, and trucks used in open-cast mining, but also turbines and generators, power distribution plants, and traffic lights. In short, automation technology is in use everywhere, in every factory, in every infrastructure facility and every building.
The connection of different devices to each other using Fieldbus systems represented an important leap forward in innovation. This development was driven by the need to reduce wiring costs. Using a single thin two- or four-wire cable, it was now possible to transmit signals within just milliseconds, paving the way for significantly faster response times. It was this development 20 years ago that laid the foundation for digital communication in the industrial environment. The world of automation, which had previously been purely proprietary, was opened up to the wider public with the advent of today’s standardized Fieldbuses.
Higher-level control systems in particular could now be more selectively integrated into the production process using bus systems. This allowed control or production planning systems to directly govern what was produced when, where, and how, starting with orders and bills of material arriving from higher-level business administration and order management systems. This type of networking has now become so established in production that almost every other machine or plant is networked on the basis of standards such as Ethernet, TCP/IP, and real-time Ethernet.
Seen in this light, automation and production are becoming totally interlinked, transforming into a whole new digital world. The individual automation networks are turning from hierarchical systems into networked systems on every level. These can not only be adjusted with extreme flexibility and timeliness to fit in with existing or future plants and assignments, but also be relied upon to transport data in real time to its respective destination, so increasing the transparency of production.
For job planning and logistical processes, this is creating a boost to productivity on a par with that experienced by the production process itself back at the beginning of the 1980s with the advent of automation technology.
Information from the digital factory
The German Engineering Federation (VDMA) and Tetragon Consulting in 2006 carried out an analysis of packaging machine productivity in the pharmaceuticals industry. The results make for sobering reading: With effectiveness averaging just 24% to 29%, existing systems are not operating under optimum conditions.
Technically speaking, although the machines used in many filling and packaging lines offer a very high level of availability, production supervisors still suffer the effects of stress. The cause? Reasons include the diversity of products, container types and variants, dwindling batch sizes, and filling quantities. These factors not only give rise to faults at the machines themselves, but also compromise productivity as a result of time spent on adjusting and cleaning as well as machine setup processes.
Other causes of stress are unplanned production disruptions, poor job planning, and inefficient processes. These individual incidents are multiplied in networked lines to create stressful situations for supervisors. Every time the unpacker, labeler, filler, packer, or palletizer is missing something, the whole line grinds to a halt. To achieve a higher level of efficiency, resetting processes have to speed up or detailed production planning has to be improved. This is only possible with the benefit of data: Which process and which product is causing the bottleneck? What is the reason for the failures? Any attempt to optimize processes can succeed only if an overarching connection exists linking every aspect of production.
Successfully implemented examples clearly demonstrate that the investment pays off. By a simple analysis of the data alone, output can be increased by up to 10%. But this requires the availability of information derived from the process. The most important key process indicator is overall equipment effectiveness, or OEE for short. OEE analyzes availability, plant performance, and quality and reduces these to a numeric value which is shown relative to the achievable optimum.
Using up-to-the-minute key indicators for capacity utilization, efficiency (volume per hour, production per hour of work, per employee), production figures (target /actual comparison, machine run time, production output, machine cleaning time) and target figures, order throughput times and inventories of raw materials or cost-intensive input materials can be optimized.
We know from experience that companies that ignore these key indicators generally work well below their actual capacity. However, determining the OEE is not quite this simple. Manually this is a highly cost-intensive process, the collected data offers little scope for statistical verification, and only retroactive conclusions can be drawn.
While the key to successful implementation is complete networking of production facilities, interlinking the thousands of components at work in production is not without security risks. Unlike an office environment where a continuously updated and standardized virus scanner works hand-in-hand with a suitably configured firewall to create intrusion protection and a demilitarized zone, industrial applications require their own dedicated security solutions. The principle is the same, but the execution calls for quite a different approach: Where plant IT infrastructures are concerned, early detection of an intrusion is paramount.
David Heinze is the marketing manager for the Industrial Automation Systems at Siemens AG. Heinze studied electronics engineering, and after a career in research and development and in product management, he now is responsible for the global marketing for industrial security for Siemens.
Case Study Database
Get more exposure for your case study by uploading it to the Consulting-Specifying Engineer case study database, where end-users can identify relevant solutions and explore what the experts are doing to effectively implement a variety of technology and productivity related projects.
These case studies provide examples of how knowledgeable solution providers have used technology, processes and people to create effective and successful implementations in real-world situations. Case studies can be completed by filling out a simple online form where you can outline the project title, abstract, and full story in 1500 words or less; upload photos, videos and a logo.
Click here to visit the Case Study Database and upload your case study.