IT in Manufacturing


Moving from business intelligence to intelligent operations

March 2013 IT in Manufacturing

Manufacturing needs to be more agile, adaptive and efficient all over the world to stay in business. The response to market needs has to be faster and delivery needs to be at a lower cost than ever before to be globally competitive. Companies are starting to realise this and as such they are looking at ways to increase the efficiency of the business by utilising the available data better.

In my opinion, companies are misdirecting their focus at the top end of the business. Increasing the speed of purchase order processing or invoice generation will not bring about major costs savings for any business, when compared to increasing the speed of product change-over on a production line. Money is continuously being invested to improve business reporting and ensuring financial transactions are captured in real-time. At a great cost.

Compared to the spend at Business/ERP level, the ratio of actual systems and solutions spend at operational levels is very small. I have seen a number of companies willing to spend R200 million to upgrade the ERP system (with no proven return on investment or ROI), but the same company is not willing to spend R2,5 million on a project with a proven ROI at factory-floor level. I believe that one of the main reasons is that the manufacturing industry focus has shifted too much towards the shareholders, to the detriment of the customers. Systems are implemented to ensure accurate and timely financial reporting to executives and shareholders, but to the detriment of ensuring effective delivery of a quality product.

Companies spend a lot of money implementing ‘business intelligence’ systems to enable accurate reporting, bringing together data from various parts of the business into an environment where data can be visualised from different perspectives. This leads to more efficient reporting and decision-making at the corporate level. The problem is that it does not help the manufacturing facility improve output, efficiency or quality.

At operational level, real-time data is needed, in real-time, to make decisions that can directly influence manufacturing operations. The focus here is ‘real-time’. Operational intelligence – OI (or enterprise manufacturing intelligence – EMI) systems can enable better decision-making at operational levels, with the corresponding financial impact. In one of my previous articles my co-author quoted some research that most processes only operate in their optimum state for 20% of the time. Increasing this to 75% or even just to 50% of the time will have huge financial impact, from processing time savings to raw material savings to yield improvements to quality improvements. Managing the state of the process can be done at control, MES/MOM or EMI levels, as long as it is shown to the operator in such a way as to influence his decisions and behaviour.

Whatever measurements are provided to operators, supervisors or managers, need to be considered very carefully to ensure that it drives the behaviour toward an effective business outcome. Take the following scenario as illustration.

Company X prides itself on customer service. They have spent several million ensuring that they meet their delivery schedules. At plant level, operations are measured against compliance to schedule. Teams that meet their schedules consistently are incentivised and those that miss their schedules consistently are disciplined. This measurement has driven behaviour and the company has a reputation for their good customer service.

During a production campaign, one of the extruder dies break. The machine stops and production cannot continue until a replacement die can be located. Luckily the production manager has established a relationship with a vendor that keeps an emergency stock of dies located 30 minutes away from the factory. No stoppage will thus last more than 90 minutes. None of the other extruders are set up to take over the job in progress. To meet their schedule, operators and maintenance personnel are moved around so that all available hands can be utilised to get the machine up and running as soon as possible. The production manager has trained his team so well that the die-exchange process takes only 75 minutes before the machine is up and running again. The operations process was efficient and everyone worked together smoothly. Although the whole team works overtime for one hour, the schedule was met and everyone is happy. Or are they?

What if the order the machine was processing was cancelled just after production started? What if the customer pushed out his delivery date by a week? Would mobilising the workforce to achieve this efficient die change-over process contribute toward the effectiveness of the operations outcome? Not in this case. A whole team has been paid overtime to meet a schedule that has become irrelevant or dated. If however, the operations were more intelligent in their use of data, they could have decided to change over to a different order or even a different product altogether. This would have increased the operations effectiveness greatly as the team would have produced to fill an existing order due for delivery the next day.

Based on the above scenario operations were very efficient, but not very effective. Making operations intelligent requires information in real-time. If the production scheduler knew that the order moved out or was cancelled, and he had the information regarding the machine breakdown in real-time, he would have changed the schedule immediately to stop the production run and change over to a different order. This is making the operations more intelligent to thereby improve the effectiveness of the operations outcome.

Consultants and people implementing business intelligence and operational intelligence solutions often lose sight of what the business is all about. They focus on streamlining business and operations processes and making them more efficient, where the goal should be on making the whole business more effective.

To indicate the point above, I would like to share an anecdote about a company I dealt with a number of years ago. The company consisted of a number of mines and a number of smelters. Transfer of material between the mine and the smelter was at a pre-determined ‘internal transfer price’. This transfer price for its high-grade material was lower than the price it could get from a competitor’s smelter. To increase the profit of the mine, the mine thus sold the high-grade ore to the competitor and sold only the low-grade ore to their own smelters.

Obviously this led to the smelters running very inefficiently from a power consumption perspective and they required a high-grade material to supplement the low-grade ore they receive from the mines. The smelters then bought semi-processed high-grade material from the smelters of the competition. This worked very well and increased the efficiency of the smelters to profitable levels again. But, in effect the smelters were buying their own high-grade material (supplied by their own mines) back from the competition (who was making a healthy profit on the deal). If this situation was known to a wider audience and debated by executives, various alternatives would have been possible that would have benefitted the company more. For one, they could have changed the transfer price so as to negate the advantage of selling to the competition smelters, giving the smelters the opportunity to process a better mix of high and low-grade ores. They could also have negotiated a deal with the competition to toll-process some of their high-grade ore to the semi-processed state, thereby paying only for the processing and not for the ore as well. Lastly, with the value they lost by selling high-grade material to a competitor, they could have built a plant to do their own semi-processing and so cut out the middle man.

From a financial and business perspective, both the mine and smelter operations became more efficient and profitable, but they did not become more effective. In effect, they gave money away.

In the final analysis, it does not help a business to automate or make any process more efficient if that process does not also make the business outcome more effective.

For more information contact Bytes Systems Integration, +27 (0)82 654 0290, gerhard.greeff@bytes.co.za, www.bytes.co.za





Share this article:
Share via emailShare via LinkedInPrint this page

Further reading:

Development of motor control units for automotive industry
Siemens South Africa IT in Manufacturing
SEDEMAC has adopted the Siemens Xcelerator portfolio of industry software, which is used in the development of its motor control units and engine control units. The motor control units are used in EVs, hybrids, ebikes and power tools, while the engine control units are used for off-road and on-road engines.

Read more...
Cybersecurity and cyber resilience – the integrated components of a robust cyber risk management strategy
IT in Manufacturing
Organisations continuously face numerous cyberthreats in today’s digital landscape, and while many prioritise cybersecurity to safeguard digital assets, their strategies for cyber resilience often become neglected.

Read more...
Sustainable last-mile delivery electric trucks
Siemens South Africa IT in Manufacturing
Workhorse Group, an American technology company focused on pioneering the transition to zero-emission commercial vehicles, has adopted the Siemens Xcelerator portfolio of industrial software as it builds electric trucks for sustainable last-mile delivery.

Read more...
South Africa’s role in the AGI revolution
IT in Manufacturing
AI has found its way into general conversation after the emergence of large language models like ChatGPT. However, the discussion is increasingly turning to the search for Artificial General Intelligence (AGI), which could entirely change the game.

Read more...
Predictive asset performance management with ABB Ability Genix
ABB South Africa IT in Manufacturing
The ABB Ability Genix APM suite is a comprehensive asset management platform powered by AI, IIoT and model-based predictive data analytics. This enables a paradigm shift towards a more proactive and predictive asset management approach.

Read more...
Intelligent automation primed for $47 billion revenue by 2030
IT in Manufacturing
According to GlobalData, the intelligent automation market is set to grow from $18 billion in 2023 to $47 billion in 2030, driven by advancements in AI, particularly the rapid adoption of generative AI.

Read more...
Chocolate manufacturing with Siemens Xcelerator
Siemens South Africa IT in Manufacturing
Freybadi, one of the largest chocolate manufacturers in Indonesia and a trusted supplier of chocolate in the Asia-Pacific, Middle East and African regions, has adopted the Siemens Xcelerator portfolio of industry software to optimise its manufacturing and production processes.

Read more...
A CFO’s guide to unlocking the potential of gen AI
IT in Manufacturing
CFOs of leading global organisations understand that their role extends beyond mere financial oversight; they are pivotal in steering organisation-wide transformation, particularly in the realm of technological advancement.

Read more...
Higher level cybersecurity certification for Schneider Electric
Schneider Electric South Africa IT in Manufacturing
Schneider Electric’s EcoStruxure IT NMC3 platform has obtained a new and higher level of cybersecurity certification, making it the first data centre infrastructure management network card to achieve SL2) designation from IEC.

Read more...
Industrial automation edge AI
Vepac Electronics IT in Manufacturing
Teguar, a leading provider of industrial computer solutions, has announced an innovative partnership with Hailo, an AI chip maker renowned for its high-performance edge AI accelerators. This marks a significant step forward in Teguar’s mission to provide powerful and reliable computing solutions for a wide range of industries.

Read more...