Enhancing operational efficiency – A case study

Managing a fleet involves multiple challenges that impact operational efficiency, and digital solutions have demonstrated to improve both the productivity and the profitability for the fleet managers. We at Intangles are consistently working with the fleet managers to infuse Artificial Intelligence based solutions into their workflow to address operational challenges of improving the up-time of their fleet. The key to achieve the objective of optimal performance involves integrating predictive based analytics and driver behavior scorecard into the fleet operator’s digital thread. The whole idea is to chart a fleet care pathway that touches multiple aspects right from building a digital twin of the engine, which is the core of any fleet to studying the interaction between the engine and its operator, in this particular case, the driver.

This paper demonstrates how a digital twin solution developed by Intangles has elevated the performance of the fleet at Purple based out of Pune, which is one of our 630+ customers sites we serve across the globe (http://www.purplebus.in/). The study highlights a complex interaction between multitude of aspects that need to be analyzed to overcome challenges that are many a times financially draining to the fleet operator not only because of the lost opportunity due to downtime but also because of the penalty clauses imposed by the governing authority by not performing optimally. In a traditional framework, Purple fleet managers relied solely on manually inspecting their fleet of buses for nearly six hours every day and ranking their fleet on operational worthiness, with no scientifically driven metric. Adding newer scan tools was not a feasible and cost effective solution as it did not capture some of the real time engine scenarios that many a times did not show up during regular service schedules. Needless to mention here is that all this led to poor fuel mileage and increased cost.


Intangles demonstrated that the performance of the fleet could be significantly improved by building an edge solution that interacts with the cloud and leverages the high compute required for building a digital twin of each engine. The team deployed edge devices across all the buses at the two depots identified by Purple. First depot (Depot A) selected was the main hub that not only served within the city limits but also serviced intercity transits. The second depot (Depot B) served local transportation within the city limits. The millions of data points generated during the period of study from the buses corresponding to both the depots were analysed using Intangles proprietary predictive algorithms hosted on the Amazon cloud infrastructure (AWS) and continuous real time feeds were made available to the fleet managers at the command center of the two depots. Simultaneously, the actionable data was also sourced to the mobile apps for rapid response so that on the fly data decisions could be taken. The predictive algorithm generated alarms for potential failures in the vehicle even before the bus reached the depot after completing the trip. This not only averted eventual failures in the field, but also ensured proactive timely maintenance and enabled efficient inventory management of spares. Vehicle Health Monitoring with prognostic alerts enabled the maintenance team to gain insights on gaps within their operations.

A typical dashboard demonstrating the health of the buses is shown in the two tables pasted below. The point noteworthy is that for one of the vehicles (MH 14 CW 4254), major fault codes were captured during its regular trip and the reasons associated with the fault codes were also highlighted for the maintenance team to plan their repair strategy.

Each vehicle shows up on the dashboard under vehicle health in order of how major or minor the issues they faced as seen in the image on the left. Information tab of individual buses gave exact causes and possible remedies to those causes.



The data analysis at command center threw remarkable insights on sub-optimal monthly fuel savings in spite of lesser breakdowns for buses from Depot B. The vehicles at Depot B though in good health, yielded poor mileage (after comparing to vehicles operating the same routes but serviced at ‘Depot A’ to get correct comparison and benchmark mileage values). The chart below shows the comparison at fleet level between the two depots.

The primary goal of reducing breakdowns was achieved by bringing it down to 0. Depot A, which looked after more buses had better performance contrary to the expectation. Depot B servicing only 50 buses performed poorly as compared to ‘Depot A’.

Interestingly, the instances and total time of idling, over speeding, hard braking & harsh acceleration were the key reasons identified for the fuel losses at Depot B when compared to Depot A. To exacerbate the problem further, it also became evident that free running (coasting in neutral gear) and incorrect gear usage were other key factors that led to significant fuel losses at Depot B. Hence, the sub-optimal performance was attributed to driver behavior, which led to a design of new driver training program.


A first of its kind, Driver Score Card and Incentive Program’ was designed to improve overall mileage of Depot B with the goal of inculcating healthy driving practices so that the benchmark set by Depot A could be achieved. The gear ratios during their driving were tracked, and the drivers were also advised to maintain a certain speed before a higher gear could be used.

The gear usage could be tracked to determine where the driver needs to improve to reach the optimum usage.

With these parameters, and consistent feedback after each trip, the drivers were able to reflect upon certain fault lines in their driving skills. Instances of harsh braking, over speeding were brought to their notice. Unusual driving practices, such as the habit of starting the bus in second or at times even in the third gear were highlighted and the consequences of doing so were informed to the drivers. A score card for each driver was generated on several parameters that also included, but not limited to external parameters, such as road conditions, load, type of bus, environment temperature, time of the day etc. One of the key aspects while designing the driver sensitization module was to keep its target audience in mind and make the whole course a learning experience even from a layman’s perspective. The performance of top drivers was incentivized, leading to a healthy competition among drivers and building a sense of collective responsibility to improve the efficiency of the entire fleet.  Taking a cue from this, the ‘Driver Scorecard and Incentive Program’ was rolled out across their entire fleet, beyond the two depots.



Based out of Pune, India, Intangles Lab is a leading startup providing Digital Twin solutions. Intangles aims to solve challenges in the industry and improve organizational effectiveness. Through innovation & excellence, Intangles aspires to be the best establishment in the field of Digital Twin which help companies meet their goals through smarter technology.

Leave a Reply

Your email address will not be published. Required fields are marked *