Predictive Analytics deals with the prediction of future unknown events. It makes use of techniques from artificial intelligence, data mining, machine learning , mathematical modeling and statistics to analyze current data from which predictions about the future are made.


The quality and representativeness of data is more important than the quantity of data collected. Hence a strategy of merely collecting and recording everything possible is often a waste of valuable resources and time. No data scientist can extract information from a dataset where no relationships between observations exist, regardless of the size of your data. On the other hand even the novice data scientist can extract powerful insights and build reliable prediction or classification models from a dataset that represents a large enough spread in variations, regardless if there was a relative small number of objects recorded.


I have the know how and experience to help you acquire the data that really drives your organization, enabling you to extract value adding business insights.


Below are a few examples on how predictive analytics can be applied to real world situations today.



Data mining




Predictive Analytics & Reliability Engineering


Maintenance is seen as the space where data mining and predictive analytics may have the biggest growth for the near future.


Data mining historical logs of running conditions and maintenance schedules may reveal unknown trends and relationships with regards to equipment up-time or failure. These can be used to make more accurate predictions towards equipment failure as a function of several factors, instead predicting future failures from merely simulating runtime for individual parts.


Accurate failure predictions provides a way of shortening preventative maintenance intervals to save on down time or extending intervals resulting in cost savings. It further aids in better just in time management of your critical parts inventory.

Intelligent Short Term Insurance

A prominent short term insurance company introduced a smart phone app in 2015 that records your driving habits. The user gets scorecard feedback and incentives to promote good driving habits. The company builds an invaluable dataset that can be used for continuously modeling their risk.


Modeling these data may enable the insurance company to classify their policy holders into smaller groups that more accurately defines any individual’s claiming risk by more than age group and location for example.


Using these models may even enable them to better categorize customers not using the app, based on alternative information sources.


Accurate classification of risk and liability will enable them to better compete on insurance premiums while at the same time increasing their profit margins.

Timmerman Analytical

PO Box 4002

Old Oak


South Africa


Phone: +27-82-335-9752

Fax: +27-86-540-4051


© Timmerman Analytical 2008 – 2017 All rights Reserved

Authorized Reseller

Intelligent Data Science


Applied Chemometrics


CAMO Software