Gary Ruck is a registered Professional Engineer and Project Management Professional in Ontario. He has over 35 years of experience in asset management and is a recognized expert in that field. He is currently the Director of Global Business Development at Deighton Associates Ltd.
Summary
Are you possibly sitting on valuable data and not aware of it? Do you have data, but you are not sure how best to use it? Every agency usually has more data than they realize and often struggles with what to do with it. Do you want to find that diamond in the rough? Read on and this blog will help you with these questions.
Introduction
Most large agencies, such as Departments of Transportations (DOTs), are often dealing with more data than they realize and struggle with how best to use it. The adage of “drowning in data but thirsting for information” rings true here. DOTs often get pitched trial data from vendors as part of pool fund studies or research and are not really sure how best to make use of this sometimes very valuable data.
The Georgia Department of Transportation (GDOT) was in this situation a few years back when one of their vendors sold them on reflectivity data for pavement markings (line striping). After some verification of this data to determine its quality and veracity, GDOT was not exactly sure how to turn it into more actionable information. Data is good, but information is better. Information allows you to act upon the data and make proactive decisions; whereas data on its own tends to produce more reactive thinking.
GDOT, at the time, was using dTIMS Business Analytics (BA) for their pavement resurfacing program, which was producing the long-term strategic works program they were looking for. After the reflectivity data was collected and delivered, GDOT and one of their in-house consultants came to Deighton to see if the team could collaboratively come up with a solution to use this data to its full potential.
The Project
GDOT, Arcadis, and Deighton embarked on a project to consume the reflectivity data in dTIMS BA and develop all the dTIMS artifacts required for a full life-cycle cost analysis (LCCA) on GDOT’s pavement marking asset. This was ambitious because this was going to be a bespoke setup built upon data that had not been used in this way before. This was a truly innovative solution with the intent of providing more information to GDOT’s decision makers than just a static dataset.
The process began first with the consumption of the static reflectivity data set that had up until now, not been used beyond just a cursory review upon delivery by the vendor. Following this, the project team met on several occasions to discuss how the work around maintaining and replacing the pavement marking asset should be done. To date, this had been done in a more reactive manner. The team developed the artefacts that go into a LCCA:
- Data requirements beyond the reflectivity data
- Key performance indicators (KPIs)
- Deterioration models
- Treatment catalogue and all related properties such as costs
- Decision trees
- Economic parameters for budget analysis
The following image, Figure 1, summarizes the data preparation process required to assimilate the various data sources together for hand-off to the analysis process. Fortunately for GDOT, dTIMS is uniquely suited for this process.
Figure 1: Data Assimilation Process in dTIMS
One of the artifacts developed is a treatment decision tree. This provides the logic for each item in the treatment catalogue on when that item should be applied in the field. This exercise on its own is a highly valuable one as it forces the stakeholders to think of the problem in a proactive, strategic way rather than a reactive one. In other words, rather than “the paint marking is worn off, we need to replace it”, the thinking is “what are the factors that contribute to paint marking degradation and what is the best approach to applying paint marking to get as much life as we can out of it”.
An example of a treatment decision tree that was developed is shown in Figure 2.
Figure 2: Treatment Decision Tree
Once the data has been consumed and assimilated and the LCCA artifacts developed and configured in the software, the strategic analysis tool is used to develop long-term condition projections and a multi-year works program by performing a budget analysis at various funding levels. The output of this analysis is a fully optimized paint marking program all based on that original unused, static data file. Figure 3 is an example of the 10-year projected reflectivity rating for two budget analyses.
Figure 3: Reflectivity Projection for Two Budgets
One additional innovation was unearthed by the project team. Previously, the pavement resurfacing program and the pavement marking program had operated independently of one another. This creates inefficiencies in that you may be resurfacing a road with relatively new pavement markings. Not only is this a cost inefficiency but a reputational one as well.
Conclusion
GDOT was able to realize the power of dTIMS by having both the pavement resurfacing and marking programs in one system, thereby allowing coordination between the two work programs to reduce the inefficiencies and produce cost savings.
This project combined valuable, unused data with innovative thinking and an adaptable software analysis tool to produce information for GDOT decision makers that previously was unavailable. Armed with this, GDOT can make more informed decisions going forward that will reduce costs, avoid reputational pitfalls, increase the value of data that had already been purchased and potentially increase the safety of the road network for the traveling public.
This effort clearly demonstrates the resulting total in this case is much greater than the sum of the parts. By providing GDOT a framework to work within, they were able to think strategically about an asset rather than reactively and unearth a diamond by doing some simple data mining and collaborative thinking.
Ask yourself, have you been involved in a pool fund study where you have received data but have not done anything with it yet? Have you recently received data from a vendor that is currently sitting in a static flat file somewhere and not being used? Would you like to see if you can increase the return on investment of that data? Are you wondering if you have what it takes to realize the same benefits as GDOT has with this project?
I am confident that most readers would answer in the affirmative and if so, the next step is yours to take. I encourage you to learn more about this project and to find out how you too can go data mining to unearth your diamonds!