For productronica 2021, ASMPT introduced the Open Automation concept, supported by white papers covering topics around the SMT production line, and introducing the concept of LEAF analysis. Automation is, however, more than just the machines in the line, as it’s part of a bigger system: the Intelligent Factory.
This article explores how modern solutions can optimise material flow of PCBs and components to release the full potential of automated SMT production lines. Based on the LEAF schema each aspect will be scored according to how beneficial automation is for it. In the summary, suggestions how to take the next step in the journey toward an Intelligent Factory are proposed.
For planning, we can consider two key aspects: the production schedule, and material management. While these can be approached separately, each complements the other to make the whole so much more than the sum of its parts.
The output of an SMT line is typically only part of the final product, making efficient, on-time delivery critical to the function of the whole factory. With WORKS Planning as part of the Intelligent Factory, many of the technical challenges of optimising a plan manually are automated, by presenting alternatives based on given priorities of volume, minimised changeovers, available production time, and more.
Material for an SMT line is predominantly packaged as reels of parts. When counted at part level, it may not be possible to know if parts are all on one reel or spread across many. Optimised placement programs may require the part on more than one feeder, or a schedule may demand the same part on different lines. This uncertainty is minimised when tracking material by packaging unit using Factory Material Manager.
ASMPT’s WORKS Software Suite is an example of a tool that can combine the strengths of scheduling and material management. When material packages are known, this can feed into the schedule planning: e.g., a part available on only one reel will only be assigned to one line or feeder at a time.
Live data consumption not only provides an accurate state of material levels but can also be used in Material Demand Calculation to predict feeder refills or new setup demands hours ahead, stretching the confidence horizon.
Material planning against a schedule ensures the right material at the right time. MFO’s material reuse feature ensures that material for a near-future setup remains on feeders, reducing teardown.
Much of the planning process can be automated, but there is always room for the imagination, experience, and lateral thinking that human supervision can be provide. The Intelligent Factory can present the current situation clearly and present the consequences of choices in a way that human factors can quickly process into the right decision going forward.
In the warehouse, material coming into the factory will require at least one manual operation, if only to remove packaging. Factory Material Manager can assist in the booking of material into the warehouse, improving visibility to scheduling systems.
When requirements for material come from the shop floor as a consequence of restocking towers, warehouse operators can quickly fill standard cassettes with material for intermediate storage in the ASMPT Material Tower. Not limited to storing reels, trays and other items can be stored in boxes too.
Once the material is in the system, from both physical and data perspectives, a lot of picking for movement can be done automatically. Where it cannot, the Intelligent Factory can advise people to minimise errors and maximise efficiency.
Delivery and Storage
The movement of material by robot is often the first thing that springs to mind when thinking about factory automation. In SMT, materials fall into three main categories:
As all three of these containers share similar base dimensions, it is entirely practical for one type of AMR to manage them all. The range of material such a solution can then manage means it will have significant uptime, and quickly deliver a good return on the investment.
Hardware solutions play their part here too: bulk storage systems such as cts’ Smart Warehouse solution for PCBs help minimise the space required for such magazines, while providing automated buffer zones, allowing the SMT production schedule to be decoupled from final assembly.
Material movement is where the Intelligent Factory can make its biggest contribution to reducing labour effort. Transporting bulky, heavy material has been the backbone of robotic automation. While SMT may not be in the same class as an automotive factory in this respect, the value of a magazine full of finished PCBAs is significant, and the risk of a manual handling error needs to be carefully considered.
As current SMT material packaging does not lend itself to automated interaction, the physical acts of splicing tapes, replacing reels or trays are the operations where human assistance gives be best return. Nevertheless, there are aspects around these refill processes that can be optimised: what, where and by when?
An intelligent factory ensures staff always have the information they need. WORKS Operations indicates the feeder that needs the most urgent attention, WORKS Preparation shows where to find the replacement material, and WORKS Operations tells when to do it.
Until SMT material packaging evolves to a more automatable form, humans will be the most cost-effective and efficient way to refill or exchange material. Intelligent Factory systems can help ensure the right material is fitted to the right place at the right time.
While the highest score went to automating the delivery of material, this does not mean it is the ideal place to start when automating material flow. A step-by-step approach should start with the path of least resistance.
Planning and material management build a measurable and controlled SMT process. Picking and refilling will ensure such tasks are repeatably completed error-free. Staff will see that automation is there to make their lives easier and build their confidence in it as benefit, not a threat. This makes the Intelligent Factory more robust and competitive.
With this foundation, the introduction of automated delivery has a much higher chance of delivering a faster return on investment, enabling the Intelligent Factory to grow and evolve over time, taking advantage of new technology, but always with the support of human creativity and ingenuity that is the key to any company’s success.
For the decision to replace human labour with machine activity, ASMPT has introduced the LEAF four-factor assessment model. Here, the production process is broken down into individual steps and evaluated according to the following criteria:
Effort of human labour (Labour)
A scale is used to assess the mechanical and/or intellectual effort required for a particular activity. Here, 1 stands for a simple, physically low-impact task, while 5 stands for work that requires a great deal of effort and/or is very difficult. Errors that can be made easily and/or have serious consequences receive a 5. The rating 3 marks the mean value or an uncertainty in the assessment.
Probability and consequences of errors (Error)
The same scale has to be used to assess how often human errors occur in a process step and how serious their effects are. 1 means: not very error-prone, with minor damage in the event of a malfunction, 5 denotes a process step that goes wrong more often - and also causes major damage.
How easy is it to automate? (Automation)
This measures how easy it is to automate a task. Factors such as cost, previous experience or interruptions can be taken into account. What is easy to automate gets a 5, processes that are difficult to automate get a
Frequency of the assessed operation (Frequency)
The aim is to determine how often the operation in question occurs, the more frequent, the higher the index. Important: the units must be consistent when comparing (e.g. the number in the hour to hour or shift to shift should be compared).
The values collected in this way are multiplied together to produce an index from which it is possible to determine whether automation is economically viable and beneficial. Individual work steps can also be compared with each other and prioritized accordingly in automation. Basically, the higher the index, the more advisable automation is.
