Predict the future

Materials World magazine
,
1 Mar 2018

Mark Hunt, CTO of OneServe, explains why 2018 will be the year manufacturing sectors embrace artificial intelligence.

The manufacturing industry is approaching a seismic shift in the way it operates. Historically, it has been a leader in technological innovations, with inventions such as steam-powered machinery and the linear assembly line defining their respective eras as revolutionary.

The fourth industrial revolution promises smart factories as one of its defining features. Although it has existed as a concept for a while, 2018 will be the year that the industry embraces artificial intelligence and, as a result, will see these factories begin to take shape.  

Artificial intelligence has been subject to media and consumer frenzy, consequently enjoying a boost in corporate investment. There is, however, a distinction to be made between the ostensible artificial intelligence that we have seen spreading across many sectors, and the technology that is genuinely exercising intelligent reason. 

The former, colloquially known as ‘ai’ (lowercase), involves sophisticated analytics and is the sort of technology that powers your phone’s autocorrect. The latter is genuine AI (uppercase) and is the technology that can make intelligent and autonomous decisions beyond the parameters defined by human programming.  

While there are currently no examples of genuine AI being used in a commercially industrial capacity, some of the biggest players have already developed smart factories that will begin to make use of it in the coming months.

Anticipating needs

One facet of the smart factory that will develop in 2018 is predictive maintenance. In this context, AI is used to predict when a machine is going to break down before it actually happens, so that it can be pre-emptively fixed, and downtime significantly reduced. The AI programmes will be able to interpret enormous and historical data sets, allowing them to identify patterns and trends that are indicative of system failures.


We will also see a similar application of AI technology informing complex business decisions. By interpreting these vast historical data sets, the software is able to understand the various functionalities and efficiencies of all sorts of things, from machinery to teams of skilled workers. Rather than making decisions based on assumptions or interpretation of ostensible facts, businesses will be able to tackle complex problems using quantified and comprehensive data provided by the AI software. 

Persistent failure of one machine can typically lead a business to routinely service or replace said machine, but the data provided by AI software can demonstrate that the failing of that machine is in fact contingent on the failing of a machine that operates before it in a chain. This will help businesses make actionable progress and encourages more efficient use of resources. 

The applications of AI within the manufacturing sector are myriad and disruptive. While ai has already started driving improvements across the sector, genuine AI takes longer to develop and implement. This year, AI used in sophisticated analysis will have a great impact in reducing machine downtime and increasingly efficiencies. The insight that these analytics give provides businesses with the data to help inform complex decisions. Smart factories will inevitably represent another revolutionary change in the way the industry operates, and 2018 is the year that AI puts in motion the steps that will lead to the eventual upheaval of current systems.