ArticlesIndustryInterlate News

Exploration Visuals & Histograms

It is time for our second article, from our series of seven, focussing on the bread and butter visualisations our analytics team use.

To receive a copy of the whole series now go to:


Critical in exploring the data and significantly useful in evaluation; insight gained from visualisations inform the whole analytical workflow. Some types of visualisations lend themselves better to one use over the other and, making the decision of which to use in each application is a learning process. At Interlate, our experience with analytics in Minerals processing plants has given us an appreciation for how to make these decisions, and we will outline some of our insights in the following series.

The following types of visualisations are predominantly used for exploration. At times with the right data transformations, they can also be used for evaluation. The exploration stage is the time to look at overall trends, such as potentially time sensitive relationships, commonly caused by, different geo-metallurgy ore bodies, or flow sheet changes. A good set of exploration graphs will ensure you are not drawing any conclusions or valuations on something that is an outlier or making any ‘ground-breaking’ discoveries such as “to improve recovery we need to increase the feed grade.”


Starting off in our series of exploration of visuals are histograms. Simple enough, they show the frequency and distribution of each variables. Great for finding out how much of your data actually lies past a certain threshold. Finding out if that operating point was actually unusual. At Interlate we use them often for deciding what filters are required to clean the data.

Good For:

  • Evaluating the variability i.e. time spent at certain setpoints, histograms of process variables can often indicate where previous setpoints or limits where.
  • Identifying distributions within groups; are there bi-modal or even tri-model distributions.
  • Discovering where the data exist and where filters should be applied, which is useful when trimming tails (removing outliers) .

Difficult With:

  • Identifying relationships between two or more variables.
  • Finding trends or changes over time
  • Data aggregation must be consistent throughout. A change in aggregation from 1 minutely to 5 minutely, for example, can greatly skew your results.

Tips & Tricks:

  • Histograms are invaluable when trying to figure out the typical range of set points in the plant, so they tend to be used as a first step to understanding a new area. Discovering the true variation in key KPIs through histograms is often surprising even to experienced team members.
  • Viewing histograms over different categories within the variable being examined can yield useful results.
    • An example of the above is to use histograms to show differences between time periods. This kind of result, however, should be used sparingly because scaling and other factors may distort interpretation. A time series may be used to explore and histograms to confirm.
  • F-tests and T-Tests are great statistical tools to give numeric values to the differences in histogram distributions.
    • F Tests give numeric value to the wideness of the distribution
    • T Tests to give numeric value to the centre-point or average of the distribution.
  • Histograms are great for collating information for your other visualisations in the following ways:
    • The bin sizes for histograms inform a starting point for your binned scatter plots or box plots.
    • A great rule of thumb for trimming tails is to filter out the end ranges of a histogram with counts less than 2% of the total.
    • The symmetry or lack thereof in a histogram can give valuable information around the biases in the control system or control behaviour.

This article series will focus on our key visuals, Interlate hope to share our experience to others and provide a robust understanding for their place in Clarofy our visualisation and data analytics application.

No software installation required and runs straight from your browser


Link ot first article from the series – Introduction: