Simafore provides tools and expertise to:
The Analytics Compass Blog is aimed at two types of readers:
individuals who want to build analytics expertise and
small businesses who want to understand how analytics can help them improve their business performance.
If you fall into one of these categories, join hundreds of others and subscribe now!
This article was contributed by Vaibhav Waghmare.
Some industry experts believe that the field of predictive maintenance is not growing to its full potential. According to one article on condition based maintenance, from a couple of years back, "the growth of CBM sales world wide, the market has been relatively flat for the past 5 years. Some industry experts peg the current growth rate of our industry between 4-6% per annum, while others see it in the 2-4% range". The article indicates that condition based maintenance (which is really an industry term for predictive maintenance), is now subject to routine cost cutting actions which typically affect other more mundane P&L items. Finally the article lists four main challenges which would need to be effectively addressed before the growth hits full potential.
When executives do not clearly understand how a forecast or prediction works, they naturally tend to be suspicious about its usage. This suspicion gets stronger if the forecasts or predictions are very good! A common challenge that one has to address when using certain "black box" techniques, for example, artificial neural networks among others, is that they are difficult to explain to non-analytics people and therefore spread doubt and confusion about the real benefits businesses can derive from analytics.
(This article was contributed by Vaibhav Waghmare)
We deal with many small businesses for whom virtualization and cloud computing would be excellent solutions, particularly for deploying advanced analytics. For example, we have come across many small manufacturing focused businesses which cannot afford expensive analytics software or dedicated server hardware to run the software, let alone an IT department to maintain all of them. Virtualization, by using open source software and pay-per-use server systems can address many of these challenges. However, there is one additional and very valid question – about data security, which deserves a closer look. This article examines security solutions provided by well-known hardware and server vendors. The good news for businesses looking to deploy cloud based analytics solution is that many solutions are available, to address their data security concerns.
The ideas behind predictive maintenance have been around for decades. In fact, almost 20 years ago, in a book called "Condition-based maintenance and Machine Diagnostics" by Williams, Davies and Drake (1994) they came out with formal definitions and the need for predictive maintenance. They defined condition monitoring simply as:
This article was contributed by Vaibhav Waghmare
"Our business is unique, and we don't think we can use ***" This is a common response we get the first time we talk to a company to explore if they are ready for analytics. The "***" is usually some analytics based process. For example, one manufacturing company insisted that the demand for their products is not predictable and therefore they would have no use for forecasting. Today, this same company is relying on monthly forecasting models to assess demand and validate production schedules. The more scientific assessment of their situation is that the demand is uncertain, but not wholly unpredictable. This scenario is perfectly suited for predictive analytics, because it is all about "converting future uncertainties into usable probabilities". The forecasting models do precisely this, by providing confidence intervals around next month's demand such as "80% confidence that product 1 will sell between 1200 and 1300 cases, and 95% confidence that it will sell between 1000 and 1500 cases".
How well do you understand your products and production processes? This is a challenge that large manufacturers are implicitly throwing across to all their suppliers. This is not a simple quiz with cash rewards at the end, but a more sophisticated strategy to weed out inefficiency. The challenge is simply this: as a large manufacturer, the original equipment manufacturer (OEM) will send out request for quotes (RFQs) to all their parts suppliers with the condition that the turn around time for receiving back the responses be days, not the conventional weeks or even months.