Connect with us


AI can rescue SA manufacturing



By Mohamed Cassoojee, Managing Director of IFS South Africa

South Africa’s manufacturing sector is under enormous pressure, contracting by 8.8% quarter-on-quarter in the first quarter of 2019, according to Stats SA. This poor performance from the manufacturing sector was one of the main contributors towards the country’s GDP shrinking by 3.2% for the quarter. Against this backdrop, it has become more critical for local manufacturers to look at ways to improve efficiencies and productivity.

As manufacturers around the world embrace Fourth Industrial Revolution technologies such as advanced robotics, the Industrial Internet of Things (IIoT), 3D printing and artificial intelligence (AI), the local industry needs to look at how it will remain relevant and competitive. One of the major opportunities lies in smart automation solutions like robots, robotic process automation and AI.

At IFS, we predict that more than half of the world’s manufacturing companies will use some form of AI by the end of 2021. AI is poised to change how the supply chain works, giving the leading organisations a significant competitive edge. We are seeing organisations around the world taking a pragmatic, realistic view on AI, hitting the ground running with targeted, project-based AI solutions.

Small, iterative projects

The problem is that not many manufacturers understand what the term AI really means. Some see AI as a costly, overarching system. And although it isn’t, the assumed complexity and cost has frightened off many South African manufacturers, many of which are bogged down with vast legacy systems and barely enough IT budget to keep the lights on. Yet the reality is that AI can be deployed in small, iterative projects with a proven path to return on investment.

AI technologies demonstrate many behaviours associated with human intelligence, such as planning, learning, reasoning and problem solving. Essentially AI gives us the ability to move beyond our natural limits and make sense of the vast amounts of data that are being continuously generated.

‘AI’ is made up of a slew of connected technologies – machine learning, automation, analytics, facial recognition and natural language processing. Each of these has its own use cases and benefits and the potential to eliminate mundane tasks. One thing these technologies have in common is their ability to achieve speed and accuracy beyond what any human being is capable of, and together, they demonstrate many behaviours associated with human intelligence, such as planning, learning, reasoning and problem solving.

One example would be LG, who is operating smart factories that employ Azure Machine Learning to pinpoint and predict defects in their equipment before any problems occur. In this way, predictive maintenance is enabled, which in turn eliminates unexpected delays, the cost of which can run into hundreds of thousands of rands.

One weapon in the arsenal

However, AI can’t simply be implemented. Before embarking on any AI journey, the business needs to identify a problem and decide whether AI is the technology to help them solve it. Businesses shouldn’t feel pressured by the fact that they perceive AI as being mainstream. The best approach to AI is a holistic one, that takes into account the business, industry and strategy Manufacturers need to consider the impact across all areas of the organisation, including HR, operations and so on, and think too about the timing of the proposed changes, not only the technology itself.

AI should only be one weapon in the arsenal used to address and solve real business problems. Ultimately, in the manufacturing sector, the goal is to take away the onerous and mundane tasks that can easily be accomplished by machines, and free up human resources to focus on core business objectives. It isn’t about cutting jobs, but increasing productivity, and getting humans and machines to work in unison.


Millennials turning 40: NOW will you stop targeting them?

It’s one of the most overused terms in youth marketing, and probably the most inaccurate, writes ARTHUR GOLDSTUCK



One of the most irritating buzzwords embraced by marketers in recent years is the term “millennial”. Most are clueless about its true meaning, and use it as a supposedly cool synonym for “young adults”. The flaw in this targeting – and the word “flaw” here is like calling the Grand Canyon a trench – is that it utterly ignores the meaning of the term. “Millennials” are formally defined as anyone born from 1980 to 2000, meaning they have typically come of age after the dawn of the millennium, or during the 21st century.

Think about that for a moment. Next year, the millennial will be formally defined as anyone aged from 20 to 40. So here you have an entire advertising, marketing and public relations industry hanging onto a cool definition, while in effect arguing that 40-year-olds are youths who want the same thing as newly-minted university graduates or job entrants.

When the communications industry discovers just how embarrassing its glib use of the term really is, it will no doubt pivot – millennial-speak for “changing your business model when it proves to be a disaster, but you still appear to be cool” – to the next big thing in generational theory.

That next big thing is currently Generation Z, or people born after the turn of the century. It’s very convenient to lump them all together and claim they have a different set of values and expectations to those who went before. Allegedly, they are engaged in a quest for experience, compared to millennials – the 19-year-olds and 39-olds alike – supposedly all on a quest for relevance.

In reality, all are part of Generation #, latching onto the latest hashtag trend that sweeps social media, desperate to go viral if they are producers of social content, desperate to have caught onto the trend before their peers.

The irony is that marketers’ quest for cutting edge target markets is, in reality, a hangover from the days when there was no such thing as generational theory, and marketing was all about clearly defined target markets. In the era of big data and mass personalization, that idea seems rather quaint.

Indeed, according to Grant Lapping, managing director of DataCore Media, it no longer matters who brands think their target market is.

“The reason for this is simple: with the technology and data digital marketers have access to today, we no longer need to limit our potential target audience to a set of personas or segments derived through customer research. While this type of customer segmentation was – and remains – important for engagements across traditional above-the-line engagements in mass media, digital marketing gives us the tools we need to target customers on a far more granular and personalised level.

“Where customer research gives us an indication of who the audience is, data can tell us exactly what they want and how they may behave.”

Netflix, he points out, is an example of a company that is changing its industry by avoiding audience segmentation, once the holy grail of entertainment.

In other words, it understands that 20-year-olds and 40-year-olds are very different – but so is everyone in between.

* Arthur Goldstuck is founder of World Wide Worx and editor-in-chief of Follow him on Twitter and Instagram on @art2gee

Continue Reading


Robots coming to IFA



Robotics is no longer about mechanical humanoids, but rather becoming an interface between man and machine. That is a key message being delivered at next month’s IFA consumer electronics expo in Berlin. An entire hall will be devoted to IFA Next, which will not only offer a look into the future, but also show what form it will take.

The concepts are as varied as the exhibitors themselves. However, there are similarities in the various products, some more human than others, in the fascinating ways in which they establish a link between fun, learning and programming. In many cases, they are aimed at children and young people.

The following will be among the exhibitors making Hall 26 a must-visit:

Leju Robotics (Stand 115) from China is featuring what we all imagine a robot to be. The bipedal Aelos 1s can walk, dance and play football. And in carrying out all these actions it responds to spoken commands. But it also challenges young researchers to apply their creativity in programming it and teaching it new actions. And conversely, it also imparts scholastic knowledge.

Cubroid (Stand 231, KIRIA) from Korea starts off by promoting an independent approach to the way it deals with tasks. Multi-functional cubes, glowing as they play music, or equipped with a tiny rotating motor, join together like Lego pieces. Configuration and programming are thus combined, providing a basic idea of what constitutes artificial intelligence.

Spain is represented by Ebotics (Stand 218). This company is presenting an entire portfolio of building components, including the “Mint” educational program. The modular system explains about modern construction, programming and the entire field of robotics.

Elematec Corporation (Stand 208) from Japan is presenting the two-armed SCARA, which is not intended to deal with any tasks, but in particular to assist people with their work.

Everybot (Stand 231, KIRIA) from Japan approaches the concept of robotics by introducing an autonomous floor-cleaning machine, similar to a robot vacuum cleaner.

And Segway (Stand 222) is using a number of products to explain the modern approach to battery-powered locomotion.

IFA will take place at the Berlin Exhibition Grounds (ExpoCenter City) from 6 to 11 September 2019. For more information, visit

Continue Reading


Copyright © 2019 World Wide Worx