IMPLEMENT PREDICTIVE MAINTENANCE FOR BEST RESULTS
data ; testing data and process data . But how do you join , clean , transform and analyse all this ? The problem is basically one of ‘ data-rich , informationpoor ’. The current methods involve data scientists spending weeks and sometimes months focusing on a specific repeatable issue and building the perfect model .” Apart from this , can anything else be done ? “ The answer is surprisingly yes . There are technologies emerging into the mainstream from a highly scientific background that can either automate or greatly assist with the joining , cleaning and transformation of disparate and unstructured data and some that have predictive analytics algorithms that can work with dirty and incomplete data . There are different ways to resolve the same problem and different strengths and weaknesses of each . Some don ’ t have the marketing muscle of the big brands , but they gain credibility from actual performance ,” said Somers .
Know your machines So the point is not about one company being somehow better than others , it is about buyers understanding what is really going on under the hood and what the limitations are . Also it is not necessarily about the technology at all but about not getting carried away with the hype , and just implementing good processes .
For instance LNS research recently published useful recommendations for manufacturers looking to implement predictive maintenance in their plants . Start by categorising assets according to reactive , preventative and predictive maintenance . Begin with the assets that would benefit best from a predictive maintenance approach . These are the ones you will want to start with , especially if you are trying to demonstrate quick wins with senior management , from whom you may require endorsement for a more comprehensive predictive maintenance program .
‘ Overmaintaining ’ can really be as bad as too little maintenance . The payback in preventive maintenance is not just trying to avoid failure ; it is also trying to avoid over maintenance , especially from a cost perspective .
So in essence , think before you make critical business decisions and ensure the software you purchase works for your systems .
13