
![]() |
Charlotte Stonestreet
Managing Editor |
Making the most of modular
25 September 2018
Niche trends and product individualisation are forcing manufacturers to deliver smaller batches of discrete products in less time, yet traditional process automation architecture is typically not designed for this type of production. Derek Lane explains how a modular approach can help
Process industries are facing uncertain times and unprecedented challenges. The need to maintain uptime and reliability, synchronise supply with customer demand, maximise available capacity and deliver smaller batches in shorter lead times has expedited the need for new technologies to keep pace. Energy and raw material prices are increasing upstream, while customers and wholesalers are demanding higher quality, faster delivery and lower prices downstream. This is squeezing margins and forcing industries to find new areas in which efficiency gains, however marginal, can squeeze more out of less.
The internet gives end customers more choice than ever before, thus allowing them to be more specific about what they want, and more discerning about what they don’t. Even if a product or range does not yet exist, there is likely to be a person or company that – for a price – can make it. The growing popularity of craft and artisan products has meant micro-companies, for example craft brewers, are increasingly having to outsource elements of production to meet demand for their most popular lines, for instance packaging or labelling. This has led to an increase in smaller production runs focused on capitalising on niche trends and demands.
Products can even be individually personalised during the production process, for instance customised clothing or even birthday cakes. Meanwhile on the supply side, the growth in digital allows manufacturers to offer much wider selections without the costs of running and stocking physical shops, as products not in stock can be manufactured to order once there is sufficient demand.
The catch with batch
So far, so good; customers and manufacturers can have their individually customised cake and eat it too. However, if every product or every few products is different, then implementing the necessary changes quickly on an automated production line can be difficult. There exists a higher chance that mistakes and errors might occur during small batch production compared to a continuous flow of one standard product. Even artisan products have to be manufactured to national and international quality and safety standards, meaning the process and final unit quality must be thoroughly tested for each product line.
Furthermore, if a bespoke product is rendered unfit for sale either on the production line or in transit it cannot simply be substituted for a similar one, it must be remade. This can substantially increase the cost of production-side errors and wastage as a whole line may have to be reset in order to remake just one unit.
It can also be inefficient. If each batch requires the line to be reset and reconfigured, then each batch effectively requires its own start-up and shut-down, as well as a potentially large amount of configuration in between. As well as taking much more time than a continuous process with no end point, constantly starting and stopping the line will also lead to greater energy use and strain on equipment. Monitoring and optimising efficiency can also be difficult when setpoints may be different with each new product.
If manufacturers are to deliver smaller and more varied product batches, new automation solutions are needed that can cope. Processes must be reliable, repeatable and error-free, with the ability to adapt on the fly and quickly ramp production up or down to match sudden changes in demand. The market is moving faster than ever, and manufacturers have to move with it, combining the agility of batch production with the speed of flow production, all while keeping efficiency high and costs low.
One controller to rule them all?
This is a challenge, as process automation architecture has traditionally involved one large, powerful master controller governing sequencing, motion and I/O. This has its advantages, particularly for continuous production with highly predictable or constant demand curves. Decades of refinement and standardisation have meant that systems are typically optimised at the design stage for a particular plant layout, specific product types and a defined level of throughput. Monitoring and optimisation becomes easier as most setpoints typically remain the same, and engineers who are familiar with the process can more easily spot and rectify any processes operating outside of normal parameters.
However, if systems are designed and optimised for one setup from the outset, this can make changing lines, or increasing and decreasing capacity very laborious. Any changes to the production sequence require detailed knowledge of the application software for all processes and sub-processes. A change to an algorithm or line of code can have unintended effects elsewhere on the line, and so for any changes made the whole line must be thoroughly tested, further adding to engineering time and costs.
Master controllers capable of operating all processes from a single point are necessarily highly powerful and therefore very expensive. A malfunction to such a controller can cause lengthy downtime, as a replacement has to be sourced, and even if spares are available the entire process sequence may have to be reprogrammed.
A modular approach
Recent technological advances have allowed process industries to tackle these problems by taking a modular approach to automation. Rather than having a single block of code executed by a master controller running a production line from beginning to end, this subdivides tasks into smaller, more manageable chunks. Control functions can be delegated to smaller, less complex and expensive controllers governing each module.
Modules are thus equipped with their own intelligence to allow them to carry out tasks with less reliance on the master controller. Parts of a process can now be inserted, removed or adapted without affecting others around it. Development time is subsequently reduced. In the event of a fault, problems can be isolated to the affected module, and then diagnosed and fixed quickly and with less disruption – in some cases while production continues around it. Maintenance teams can have spares ready to go for critical modules, so that an existing programme can be dropped in from the back-up library or tweaked to accommodate new hardware. Upscaling is also easier, as entire pre-tested control subsystem programmes can be dropped into other controllers. This is clearly far quicker than writing new code from scratch.
A modular architecture still uses a central controller, but it needn’t be anywhere near as powerful since control functions are distributed to smaller, less expensive controllers governing each module. The price of small PLCs and PACs is coming down all the time, so this represents a significant opportunity to improve agility while reducing costs by effectively delegating control and monitoring functions to each module. It also offers the potential to send more data from each module back to the control room than a single master controller might typically be able to.
Innovative automation solutions
Controllers and I/O systems have had to evolve to meet the requirements of modular automation. WAGO’s SYSTEM 750 can be almost infinitely scaled up simply by adding more I/O slices to a node, without disrupting existing processes. The WAGO system’s compatibility with all commonly used fieldbus protocols allows it to operate effectively alongside other brands, while a drag-and-drop programming interface allows modules to be quickly configured and reconfigured to adapt to different product and process requirements.
Equipping modules with their own intelligence is achieved through WAGO’s DIMA system (Decentralised Intelligence for Modular Applications). This was developed in response to NAMUR’s 2013 NE 148 recommendation that automation solutions for modular process systems clearly divide functions between module controls and higher-level control systems.
DIMA divides system architecture into two areas: the management level, which assumes high-level process monitoring and procedure control, and individual process modules, which handle all base functions. Module Type Packages (MTP) describe the modules, and contain information on process control integration, the functions of operation monitoring and batch functions. The module supplier configures the module while the system operator uses the appropriate MTP in integration engineering before configuring the overall system, eliminating the need to plan and programme the automation system manually while reducing the likelihood of errors. Communication between management-level and module is through service-oriented architecture. According to WAGO, modular system construction can reduce development time for process technology systems by up to 40%.
The bottom line
Modularisation is not a brand new concept – the desktop computer, in which similar parts can be swapped in and out with standardised compatibility – is an example of a modular system in action. Even so, many companies are still using centralised automation systems and potentially missing out on some of the cost-saving, efficiency and productivity opportunities. Modular automation equips companies with the ability to react instantly to changes in demand or fashion, and ramp production up or down quickly and easily, significantly reducing operating costs.
For industries like food and beverage, where recipes and ingredients frequently change yet homogenisation and end quality are essential, a modular approach can allow food processors to add, remove or change recipes from a function library, with almost no interruption to production. Modular machinery can also be easier to clean, as processes that aren’t required for a particular production run can be stopped and maintenance carried out without necessarily having to shut down the whole process. This is ideal for industries such as medical supplies, where regular washdowns are essential to prevent contamination.
Even if an entirely modular approach is not necessarily a perfect fit for every production line, almost every industry could benefit from modularising discrete processes where possible, reducing reliance on a single point of control. If production no longer needs to be paced by the slowest operation then modules can be optimised individually, which is far quicker and easier than having to reprogram and extensively test an entire process from a central controller.
In essence, modular automation is about creating several plants within a plant, making each segment of the production process more manageable, more flexible and cheaper to run. The idea of infinite granular customisation, whereby each individual product can be fully tailored to the customer’s particular preferences, and instantly manufactured and dispatched from the same automated product line one after another, is now technologically feasible using modular solutions from companies like WAGO. Such systems require careful design, but afford unprecedented agility to react quickly to market changes while keeping wastage and downtime to a minimum.
Derek Lane is WAGO automation manager and PI UK deputy chairman
Key Points
- If manufacturers are to deliver smaller and more varied product batches, new automation solutions are needed that can cope
- Systems designed and optimised for one setup from the outset can make changing lines, or increasing and decreasing capacity laborious
- Recent technological advances have allowed process industries to tackle this problems by taking a modular approach to automation.
- The terminator
- Streamlines commissioning
- Screwless connectors & slimline power supplies
- Systems & machines controlled, monitored & maintained from anywhere in the world
- Space saving controller
- WAGO expands robust automation system for extreme environments
- Three ways to wire
- Food industry innovation award
- Remote control & monitoring
- Component supply deal
- No related articles listed