New Technologies for Old Processes: Manufacturing Optimisation

There is much coverage of the current trend towards integration of manufacturing processes and their sensors, control software and communications into larger networks, and the resulting gains in efficiency and productivity. Many new manufacturing and process facilities are incorporating this functionality into their equipment and processes. Meanwhile, owners/operators of legacy manufacturing processes rely on experienced operators who know and understand their machinery thoroughly to maintain process throughput and quality. However, is this really sufficient (and cost–effective) in an increasingly competitive environment, and are there real performance and financial gains to be had from improving the monitoring and control of key parameters? In this article we consider how operators of older (and perhaps not so old) production plants can potentially benefit from some of the latest innovations in sensing and control. 

As in many aspects of life there is a spectrum of technology approaches to process control in industrial and manufacturing processes. At one extreme, modern machinery is highly automated and well characterised. At the other, many older mechanised (continuous) production processes – typically manufacturing lower cost or commodity products – still operate under manual control.  It is processes such as these that stand to gain most from continuous monitoring.

Often, a process control / monitoring regime is designed around what was available at the time a plant was constructed or a production machine was installed. Technology is evolving continuously, and measurements are becoming available which, when applied appropriately, could have a dramatic effect on understanding and control of a production process. The challenge for the owner/operator of a plant is how to identify the innovations that could bring real economic benefits.

One solution may be to install an off-the shelf sensor and control system (probably provided by the manufacturer of the production equipment). However, if there’s a compelling reason (such as poor product yield or process control) to measure and control a parameter or machine setting, and a suitable sensor isn’t available off the shelf, then the next question is ‘why hasn’t this been done before?’ It may just be that no-one had thought of it, or there are obstacles such as:

• Adding a sensor is perceived as being too expensive.
• A competitor has a patent which blocks development.
• Making the measurement isn’t technically feasible.

If this is the case then it can often be cost-effective to develop a novel solution, which could be a ‘me too’ sensor but with reduced costs, or a ground breaking accuracy increase, or indeed a novel combination of existing sensor approaches with some additional data processing/fusion as a proxy for the desired measurement. This is the type of challenge we relish at Cambridge Consultants.

Selection of the sensing approach calls for much more than a scientific/engineering process and involves a deep understanding of the economic realities involved, along with the measurement and operational/maintenance requirements and the environmental conditions that will be encountered. One example of where we have recently applied real innovation is developing optical sensor combinations for monitoring critical process settings; this particular case went on to reveal new manufacturing issues (‘unknown unknowns’ as it were) of which the operator was unaware; it’s often the first application of a new sensor to a process that yields the most unexpected and valuable information. Another recent example involved concepts for non-contact measurement of the mechanical strength of a product to provide early warning (in real time) of changes in quality as it moves through the production process at full line speed. In both these cases, the implementation of these measurements could bring the operator greatly increased profitability through improved product yield and, because of the improved process control, the ability to reduce the amount of excess material designed into the product. This ability to innovate whilst addressing practical business requirements is key to bringing commercial success (and a healthy return on investment) to this type of development.

Of course, data from sensing alone clearly is not the answer and the actionable information derived from sensors only has value if acted upon and if real improvements in costs, productivity and/or quality are realised. Used appropriately, information can be priceless – even more so if you can get it ahead of the competition. It can enable more efficient control over a process, or give a high level of confidence that your product is at an optimum. Best of all, it can give a headline feature which differentiates your product.

If you are the operator of a manufacturing process and are seeking improved product yields and reduced costs but can’t find a solution, if you are a manufacturing equipment provider looking for an edge over a competitor, let me ask you this: what would you really like to measure and control?