Keeping it simple in data analysis

Modern asset managers or owners will be very familiar with data, and how important it is for accurate, timely and effective decision making.  

Accounting for variables, using the right tools, and implementing specific data capture and analysis processes is important and relevant behind-the-scenes, however at times too much detail can be a distraction from the ultimate goal.

Sometimes to decide on an issue it’s enough to know if a road has high or low traffic, rather than knowing the exact number of standard axle loads. The key is to know when it is enough. Our analysis effort and decisions need to be proportional and scalable.

Data doesn't need to be complex, but it does need to be robust

There are myriad eye-candy like outputs that can be produced based on data analyses, which may include historical trends, current condition gap analysis, or forecasted conditions, to name a few.

For an asset manager or owner, however, there is really just one question needing a clear answer: How do the analysis results presented enable better maintenance planning decisions?

Take for example the desired outcome for a major roading asset owner - that users can get to where they need to go safely, efficiently and comfortably.

Of course the specific network assets, available budget and unique customer must all be considered as part of the process, but ultimately what matters is that the final decision is a reasonably right decision.

Processes do not need to be perfect, but they do need to be practical and reasonable for the end user

There’s a lot that can go into defining a problem statement and the analysis and reporting that goes into solving it – you don’t always see (or even need to see) the complex data analysis and framework behind the scenes to prepare the recommendations.

The challenge for an analyst or engineer is how to present this complex data in the most simple way.

The way results and charts are presented in reports can look impressive and highly sophisticated (read intimidating) due to the many intricacies involved with the complex solution framework.

There may be many variables, and at times some of them will be estimated.  In analysing the data containing these variables and assumptions, much effort can be put into looking for a perfect outcome.  However, to ensure a positive experience for the asset manager, the focus should rather be on producing a reasonable product which can add value to the decision-making process. 

While it may be aesthetically pleasing to present a range of complicated charts and tables, at the end of the day all this added complexity can end up being of less practical use.  

The continuous improvement action should be to simplify the process and the outcome. It really doesn't need to be too complicated

Naturally the question that follows is: How can we achieve this?  How do we formulate the results from a technical process in such a way that a reader can resonate with what is presented?

We need to stand back and let our mind simplify things. Let's find the tipping point where the information we work with is reasonable enough to make a difference and to provide valuable strategic advice to an asset manager (without of course losing the required engineering in the background).  

For each project there is a point where the technical complexity is just enough to provide a solution that is fit for purpose.  We don't always need to overthink things. The phrase “Less is more” is quite appropriate in this instance.

A few simple ways of doing this come to mind:

Be transparent and consistent with the parameters used throughout the solution process and reporting

• This enables easy communication between the analyst and the reader. By creating a common reference point used in the process, it becomes easier to join the dots between the problem statement, the analysis and the reported outcomes.

Aim for simple presentation of results

• A discussion is much more easily established with a common and simple point of departure. This also allows parties to explore the complexity together, resulting in more effective knowledge transfer.
• A simple but well-thought-out bar chart can be as effective for the asset manager, if not more effective, than a more complex statistical chart based on the same data. This allows the use of simple terms such as ‘Very Good’ and ‘Good’, as opposed to explaining statistical metrics.

Do not underestimate the importance of the initial ‘sit down’ to talk through the process and outcomes

• A safe zone needs to be created where assumptions and fundamentals can be discussed, leading to knowledge transfer and improved understanding.

Let’s take this back to our focus - to serve our customers, the road users, and enabling them to get where they need to go safely, efficiently and comfortably. 

A high volume of data can be collected, analysed and communicated in many ways.  We should continuously be asking ourselves if we are on track in serving our customers, and meeting their goal?  Are our processes efficiently supporting the maintenance planning decisions at hand? 

What we must do is avoid the temptation towards complication, and really, just keep it simple.