Tuesday, February 26, 2013

TAKING A RISK-BASED APPROACH TO SUSTAINABILITY AND OPERATIONAL EXCELLENCE IN MANUFACTURING

Manufacturing risks extend beyond
the boundaries of the site

Risk management tools are employed in some shape or form on most manufacturing sites. The two most common areas in which they are used are for safety and plant maintenance purposes, where risk- based approaches are used to identify potential failures and devise mitigation strategies to prevent these failures from occurring. In theory, the development of integrated quality management systems encompassing safety, food safety, product quality, environmental issues and more recently issues such as energy efficiency should involve some kind of risk assessment. In fact, a risk assessment should be the foundation of such systems. 

For reasons I refer to below, many organisations tend to gloss over this important step and their risk registers are not as comprehensive as they should be. They tend to identify risks through some kind of brainstorming process, which is not in itself a bad approach, but can lead to a “high-level” assessment of operational risks if the process is not sufficiently focused. The problem in manufacturing environments is that some very large risks tend to have very small origins – the v-belt that is not tensioned correctly and catches fire, burning down the facility; the valve that passes and contaminates large quantities of food products; the extraction fan that is never switched on and increases long-term occupational health problems for employees through solvent inhalation....I could go on. While management teams can and must identify strategic risks, operational excellence is largely about small details.

One of the major impediments to rigorous risk assessment is that it is resource-intensive. In the maintenance environment, Reliability Centered Maintenance (RCM) is an example of an approach that is shunned due to the amount of resources required to partition a manufacturing plant to the required level of detail, identify failure modes at the component level and then devise maintenance tasks to prevent failures using decision trees. However, without making this investment in time upfront, potentially catastrophic failures can go unidentified, and while the maintenance programme may be improved after such failures occur, the costs incurred in learning lessons in this way can be very high.  

A second challenge for those wishing to identify risks comprehensively is that this requires subject matter expertise. If a group of people spends a considerable amount of time conducting risk assessments but lacks this expertise, the chances are the risk assessment will have serious shortcomings. So resources will ultimately be committed but wasted, leading to disillusionment with the process when unforeseen incidents occur. It is therefore vital that significant investments are made in technical training and coaching, in order to equip employees to participate productively in risk assessment events.

By now you can see where I am going with this. Comprehensive, rigorous risk assessment is an essential element of any strategy aimed at sustainable, stable, continuously improving operational performance. Of course, risk assessment is only the first step, solutions still need to be developed and implemented to mitigate each individual risk in order to realise the benefits. The philosophy is simply that if we can identify and mitigate every risk, we will achieve excellence. This is in essence a lofty goal, since in practical terms, we will never be able to prevent every potential incident. However, if we try our best to eliminate every risk, those that remain will be small in number, and can be handled as they arise in line with PDCA. If we chose to deal with every risk after the fact with no risk identification upfront, we would be fire-fighting, and if your facility tends to operate in an unstable fashion, chances are your risk management practices need review.

The machinery through which risks are mitigated is comprised of the various management systems in place. Quality Management Systems specify overarching policies, how manufacturing process units should be operated, the parameters to be measured, reporting, corrective actions and the like – essentially everything that needs to be in place to ensure that excellent product quality, high levels of safety, responsible environmental performance and other key objectives are realised. These are however not the only systems through which risks may be managed. Preventive maintenance programmes are a vital component of risk management in the manufacturing environment. Human resource risks also require serious consideration, and require standardised and rigorous processes and standards for their management. Cost control procedures are a further example of tools employed to manage risk. Many of these supporting systems reside in IT platforms. The organisations that are best at developing and implementing quality management systems integrate these disparate systems into the overall quality management system through explicit linkages.  In general, the greater the number of unique systems you have to integrate, the more difficult the task, and if you could build an integrated system from the ground up you would have the ideal management system.

The complexity of quality management systems and the other programmes manufacturers may be implementing at any point in time (such as continuous improvement programmes for example) can lead to bureaucracy and confusion. In many cases a fixation with the system rather than its efficacy means that results are erratic and do not exhibit sustainable improvement. I am not knocking quality management systems or continuous improvement programmes, both of which are important vehicles for the achievement of operational excellence and sustainability. I do however believe that there is a need for manufacturers to “get back to the basics” insofar as obtaining and harnessing a fundamental knowledge of their operations is concerned. Risk assessment provides an ideal vehicle for doing so. It does however mean examining physical and business processes in minute detail, and yes, this is time consuming and requires a lot of skill and knowledge to do effectively. Once this has been done, the mitigation measures developed, if implemented rigorously, will however go most of the way towards solid, repeatable operational performance. Quality management systems provide the ideal vehicle for execution of these mitigation measures. Continuous improvement programmes require this sound foundation to be effective. Risk assessment therefore lies at the heart of the well-oiled manufacturing machine, particularly if the same approach used to identify risks is also used to unearth opportunities.  In a future post I will give you an example of a detailed, integrated, process-level risk assessment that will illustrate how powerful this approach can be as a platform for operational excellence and sustainability.

Copyright © Craig van Wyk, 2013. All rights reserved

Tuesday, February 5, 2013

VARIABLE SPEED DRIVES AND THEIR USE IN RESOURCE EFFICIENCY AND CLEANER PRODUCTION

Forced draft fans in dry cooling devices
like the one pictured can be fitted with
VSD's  to allow them to  vary speed in
response to climatic conditions or
tower performance

The speed of an induction motor is a function of its number of poles and the frequency of the power supply. Variable speed drives (VSD’s) – also called variable frequency drives (VFD’s) – are used to vary motor speed by changing the frequency of the alternating current and voltage applied. They have many uses in industry, which is to be expected given the wide range of potential applications for continuous speed control within industrial processes.

VSD’s may be controlled through the intervention of a plant operator, using an interface such as a simple button/knob which can be used to increase and decrease speed, or through a SCADA system, which can be used to manually changed the frequency output. Generally however, they are used within the context of a control loop, which varies speed automatically in response to an output from some other measured variable. For example, pump speed could be varied in response to the output from a flow meter in the pipeline carrying the fluid being pumped. Alternatively, frequency setpoints could be written into control software based on the status of the process being controlled. For example, agitator speed could be set to a minimum value while a vessel is being filled, with the speed increased after filling has been completed. There are an infinite number of ways that VSD’s can be employed in control schemes such as this. Just a few examples of common applications are:
  • Changing the speed of agitation in a mixing vessel
  • Changing conveyor speed on a production line
  • Changing the speed of a pump (and hence the flow) in fluid handling
  • Changing the speed of a compressor
  • Changing the speed at which racks are lifted from baths in electroplating
  • Changing the pressure inside a boiler (on the fireside) by changing the speed of FD and ID fans.

So far I have discussed how VSD’s give us more control over industrial processes, but what then is the contribution of VSD’s to industrial sustainability? For one, improved control means reduced rework, which translates into reduced consumption of energy, water and materials. VSD’s are however probably more recognised for their growing use in industrial applications on the basis of the energy savings they provide. Simple examples include their use to control poorly loaded screw compressors (minimising “off-load” losses), and their use in controlling the speed of fans (where flow is directly proportional to speed, but power varies with the cube of speed). Controlling fluid flow by manipulating pressure drop (e.g. the use of a control valve or damper in conjunction with a pump or fan running at a fixed speed) consumes far more energy than that required were a VSD used to control pump or fan speed instead. VSD’s also provide exceptional “soft start” capabilities (while using less energy than traditional soft starters) and can also be used to control torque characteristics and to boost torque (e.g. at start-up).  While there are losses incurred in using a VSD (of the order of 1 – 3%) the efficiency gains at the system level arising from their use generally far exceed these. As far as energy efficiency projects involving induction motors go, I tend to find many more viable opportunities through VSD applications than I find in areas such as motor replacement.

VSD’s are however far more versatile from a sustainability point of view than simply being an electrical energy efficiency tool. Their use in material usage reduction can yield spectacular results where correctly applied. Consider the following drivers of material usage in the industries in the table below, and how VSD’s can be applied to improve material usage performance:

INDUSTRY
PROCESS
DRIVER OF MATERIAL USAGE
HOW TO EMPLOY A VSD
Drum manufacturing/reconditioning
Spray painting
The speed at which drums are rotated during spray painting is an important driver of paint thickness.
Install a VSD and allow the operator to manipulate drum speed to control paint thickness, along with other variables such as solvent ratio, air pressure and nozzle design.
Powder coating
Curing
Curing time and temperature affects curing quality and rework rates
Install a VSD on the conveyance system to permit variations in residence time in the curing ovens for items of different dimensions. To be used in conjunction with oven temperature profile.
Waste management
Oil recovery in a plate separator
Residence time in the plate separator impacts on efficiency of oil recovery
Use a VSD on the supply pump and vary the flow based on the composition of the incoming effluent, thereby changing the residence time. A turbidity meter in the separator outlet could be used for automatic feedback control.
Batch chemical manufacturing
Material dosing
Accuracy with which individual chemicals are added to a batch
Use VSD’s on dosing pumps (liquids) and screw conveyors (solids)  to vary dosing rate, slowing down the flow as the target value is reached and eliminating overshoots

How else could VSD’s be employed? The answer is really that the opportunities are limited by your creativity. There are few manufacturing processes in which speed is not an important variable, and the first thing to understand is: “what is the impact of speed on the process being analysed?”. Armed with this understanding, the next question would be: “does a VSD, with its ability to provide speed control over a continuous range, provide leverage in terms of reducing resource consumption and minimising pollution?” In my experience, the answer is very often a resounding “yes”. Your final consideration would be whether this leverage/benefit justifies the investment in the VSD. In real terms, VSD’s have become cheaper over time, despite gains in the features they offer. Combine this fact with the significant benefits they can provide, and they are often easy to justify from a financial and environmental perspective.