Case Study – Qatar Survey Analysts Automate Roundabout Counts and Reduces Labor by More than 85%

Qatar Survey Analysts (QSA) is a Civil Engineering firm that collects traffic data and produces comprehensive reports for their clients. Roundabout counts are a significant portion of their business. Historically, roundabouts were manually studied by counting movements from each approach, then used to estimate the unknown movements. Since automating roundabout studies with Miovision, QSA has increased data accuracy and overall productivity.


Qatar Survey Analysts and Traffic Consultants (QSA) is a leading Civil Engineering firm in the business of traffic forecasting, urban design and land development. QSA is headquartered in Doha, Qatar, a Country that has experienced massive population growth in recent years, with tens of thousands immigrating each month. As a result, Qatar is undergoing a country-wide expansion to its transportation network aimed at relieving the congestion on its roads and improving the safety of travel throughout the country. Qatar extensively uses roundabouts throughout the country and the majority of traffic engineering projects will include studying one or more. Roundabouts are used because they are typically a safer alternative to standard intersections, with 40% fewer vehicle collisions, 80% fewer injuries and 90% fewer serious injuries and fatalities1.

‘”We believe that accuracy has the greatest impact on our success. Miovision is more than capable of giving accurate data, specifically from roundabouts.”‘ Franklin Abapo, Qatar Survey Analysts


QSA collects roundabout counts to be used in client reports for the majority of projects that they undertake. Typical roundabout configuration prevented QSA from using the established manual counting method of tracking a vehicle from origin to destination. As a result, roundabouts were manually counted by placing an observer at each approach to count simple point observations: entry, exit, circulating and next turn volumes. The reports are then combined and the movements are calculated using the ALGebraic Solution (ALGS) method of roundabout estimation to solve for the 16 unknown origin-destination Turning Movement Volumes (TMV) from each leg. All computations are made in a spreadsheet program manually and then data is partitioned into 15 minute bins. Employing four manual counters, QSA is able to conduct one roundabout count per day with basic vehicle classification.


The largest challenge for QSA was study accuracy. Manual counting was assumed to have a 5% – 8% observation error rate which, when propagated using the ALGS method to estimate TMV, can mean a low overall study accuracy rate of just 70%.

Secondly, QSA’s high use of time and resources to collect and process studies were costly and could otherwise be used to generate a better return on time invested. The manual data collection process for a 9 hour roundabout study required 36 total hours of manual counting, plus 2 hours of internal report generation.

Finally, the time-consuming nature of producing roundabout studies caused a delay between winning a project and collecting the data to begin engineering work. They delay sacrificed productivity and created a backlog.


‘”We expect our client reports to be of the highest quality. Miovision helps provide us the data we need so we can prepare materials and impress our clients well within deadline.”‘ Franklin Abapo, Qatar Survey Analysts

To overcome their challenges, QSA made the decision to automate their roundabout counts using Miovision’s automated traffic data collection system. Automating significantly reduced the amount of time and resources needed to collect data and produce reports while increasing the accuracy of the studies.

Miovision automates large occluded roundabouts by collecting video at strategic locations around the roundabout with Video Collection Units and then processing recorded video through Traffic Data Online. Miovision uses an in-house roundabout estimation method called Quadratic Programming ALGebraic Solution (QPALGS) to process measured movements from the recorded video and solve for unknown movements. The method is shown to be superior to other roundabout TMV estimation methods. QPALGS is an extension to the ALGS and O-D Trip Matrix methods of TMV estimation.

Using Miovision’s system to automate roundabout studies provided QSA with consistent and accurate results, which was previously not possible with manual counters. The QPALGS method is more accurate and more resistant to observation error than the ALGS method (see figure 1).

Figure 1: Error Propagation due to Observer Error for a high volume2 roundabout. Miovision’s average observation error is less than 4%, generating reports found to be 92% accurate.

For QSA, high accuracy and consistent results are their competitive edge. Data generated with Miovision is found to have an average accuracy of 92%. The engineering firm is the first in Qatar to upgrade their data collection equipment to high-tech automation, and this innovation has placed them ahead of their competitors, winning them business.

Using Miovision in place of manual counters at roundabouts requires 5 hours of time per study for travel and equipment set-up, opposed to 38 hours of labor when done manually, reducing labor by more than 85%. By streamlining the data collection and report generation process, QSA is able to significantly reduce time spent at each count site and eliminates the need for internal report generation. The corresponding video record allows traffic engineers to analyze behavior and backups in the roundabout, not possible with manual counters.

Automation helps QSA get studies done earlier, improving their internal velocity and leaving time to produce top-notch reports for their clients. QSA uses eight Miovision Video Collection Units to automatically collect data at two roundabouts simultaneously with less than 2hrs work. Uploading video to also relieves internal staff of report generation, saving an additional two hours of labor per study. The result is labor reduction by more than 85% per roundabout, enabling QSA to repurpose their internal staff and collect data more quickly than was possible with manual counters.


Figure 2: Error Propagation due to Observer Error for a high volume2 roundabout. Miovision’s average observation error is less than 4%, generating reports found to be 92% accurate.

QSA was able to significantly improve their process for counting roundabouts, creating more business opportunities and increasing internal project velocity. Automating Roundabout Studies enabled QSA to:
  • Generate studies found to be 92% accurate
  • Reduce labor by more than 85% through eliminating the need for manual counters to enumerate roundabouts
  • Conduct two roundabouts simultaneously, doubling data collection productivity
  • Guarantee consistent data collection with low observation error for current and prospective clients
  • Create more lead time to produce client facing reports, maintaining high brand equity for their business

Using Miovision helped to position QSA as a leader of innovative data collection in Qatar, which has helped to win business and uniquely position them within their market.

1. IIHS, Vol. 35, No. 5, May 13, 2000
2. 45 Cars per 15 minute bin