 If all the railway tracks in the world were joined end to end, we would have three different track-routes between earth and the moon. Around the world, huge investments have gone in building such a mammoth infrastructure base and the annual expenditure to maintain these resources is staggering. We spend more than $10 billion annually just to maintain the global network of tracks. Increasing the complexity and costs of this process are the various parameters involved in decision-making—from engineering evaluations to business-level constraints to safety levels. This scenario provides a huge opportunity for any IT-enabled solution— either a product or customised implementations— that could help rail road companies with efficient and effective resource allocation and decision-making.
Track Maintenance
Railway tracks form a very critical part of any rail company’s asset base, as tracks provide the company with the required business operatibility. The need for safety and maintenance of this valuable resource cannot be overstated. Track maintenance refers to the comprehensive set of activities involved in ensuring that railway tracks meet the required safety and quality standards. This includes inspection, track data collection and possible renewals. With increasing pressure on rail operators to increase operational efficiency, track maintenance is required to be cost effective,too.
The American Public Transportation Association (APTA) has discovered that nonvehicle maintenance expenditures (almost half of which are just track maintenance) form nearly 9 per cent of total operating costs, and 78 per cent of this cost is just labour cost. Such a high level of labour expense contribution and the repetitive nature of the job strengthens the case for automating the process. As compared to vehicle maintenance, track maintenance is a rather complex activity due to the geographical spread of the asset. Unlike vehicles, which can be brought to sheds or other common points for inspection, the inspection, repair or data collection on tracks requires physical movement of man and material, adding to the cost and time involved in the task. Technology has been deployed in areas of data collection, inspection and track upgrading.
An understanding of the individual sub-tasks involved in track maintenance would help identify the possible options to building solutions. A very simplistic depiction of the complex web of various sub-activities and decision points is shown in Figure 2.
IT in Track Maintenance
Most advancements in terms of automation and mechanisation have been in the sub-activities of:
Track data collection and inspection
Renewal and up gradation activities such as ballasting, levelling, tamping machines.
Most rail companies in Europe, North America and Australia have special cars that collect track-related data and also aid high-speed visual inspection. These high-speed recording cars (which reach speeds up to 200 kmph) are fitted with sensors to check track alignment, track surface conditions and the like. In certain cases, they also have an onboard analysis system that makes the task of data-collection and analysis a real-time activity. Many software solutions like DynaTrack, RailScan and ORIAN are available in the market to support the data collection and analysis phase.
Although this could represent an obvious reduction in labour costs, a lot of potential cost savings are missed on account of unplanned activities. Such speedy data collection and inspection helps in ensuring a low reaction time for maintenance, but the overall process still remains very reactive and unplanned.
The need of the day is a system that could help the roadmasters—the people in charge of tracks—identify future maintenance needs with an engineering solution and could help them plan maintenance activities accordingly.
Solution Space Map
A product mapping of available and potential software solutions that aid and assist track maintenance (based on the level of functionality) is displayed in Figure 3:
| Figure 3: Solution Space Map |
|
DSS based on
current track
conditions. Not very
beneficial, just a
hard-coded
knowledge base of
road-master
|
Ideal DSS
Systems need
very high
reliability in terms
of predictive
capabilities
and business
level engines |
Supports simple
data collection
and analysis.
Cannot plan
maintenance
nor can
generate business
level options |
Presents track
data for future
scenario. Roadmaster
must
use own
business knowledge |
One parameter of segmentation in the matrix is the predictive capabilities of the solution. Such capabilities would help the maintenance managers predict track conditions, forecast maintenance needs and thus plan and predict resource allocation. The other parameter is the one that refers to the domain knowledge that any rail operator would have acquired with the experience in the activity of track-maintenance. Such knowledge is essential to judge the resource requirements based on the physical and technical track information available. A system or solution with an inbuilt engine to trap such business cases would essentially act as a Decision Support System (DSS). To build such a DSS requires that the business-knowledge be matured and acceptable at a common industry platform, so that the reliability and the user confidence are high.
Industry Analysis
With the above framework in mind, lets look at he challenges facing the development and eployment of such Track Maintenance anagement Systems (TMMS). The ideal point to start would be to have a brief look at the trends in the industry. Rail operators need to maintain certain minimum levels of safety standards as prescribed by the respective regulatory authority and the rail companies themselves traditionally supervise the whole gamut of activities. A few rail companies did experiment with the format of outsourcing the task to third party contractors, but the experience has been so dissatisfactory that the model is almost shelved now. At the most, few vendors take small contracts to replace, upgrade, maintain or test tracks. A few possible reasons could be the criticality of condition of the tracks for profitable operations and also the level of costs involved. In the absence of sophisticated and assured delivery from third party vendors, rail companies prefer supervising the activities on their own. The operating efficiency of rail operators has been under question and pressure due to increased challenge and competition from other players not only in the rail segment but other segments of transportation. This means that rail operators need to spend each rupee in the most effective manner and hence we would be tempted to assume that the market would be eager to have a DSS.
A DSS would help the managers and roadmasters evaluate (financially and technically) and choose between the various possible options (generated by the system) ensuring that the operational efficiency sees improvement. But the business-level knowledge, as understood by accepted norms of maintenance activities for given or predicted track conditions, has not really evolved. There is still no consensus and confidence in the ‘suggested’ maintenance options for a given set of conditions. Moreover, IT sophistication has been very low in this sub-segment of the rail industry. Other rail activities like signalling and passenger booking have seen a drastic adoption of IT, but track maintenance has at best been a low preference area for IT solution deployment.
All these trends lead up to one conclusion— the market is not really ready to adopt a DSSbased track maintenance management system. However, roadmasters would like tools to predict track scenarios so that they could use their own judgment to plan various maintenance activities. Thus, the kind of solution currently desired by the industry would be an engineering solution, which uses the track information database to predict future values of track parameters.
One such solution is the Track Predictive Indices (TPI), which has been co-developed by TCS and BNSF Railway (formerly known as the Burlington Northern and Santa Fe Railway), an American freight railroad company. TPI predicts the values of five track indices and creates various reports both at the user and managerial level. The biggest challenge in maintaining assets is predicting future needs for capital and operational expenditures, and TPI helps in doing just that.
Segmentation
The various rail companies form a motley group in terms of their levels of IT implementation sophistication and this implies that a blanket marketing or product development approach might fizzle out. It might prove to be overkill for small and inexperienced rail operators and too little too late for the larger and more IT-enabled companies. Instead, it would be better if IT vendors were able to identify and classify customers according to their IT sophistication and extrapolate the specific solution that would make most sense for each particular segment.
Future Beholds
Since the present market dynamics are not suitable for a DSS with predictive capabilities, the industry needs to gradually mature in its level of IT sophistication. One approach would be to bring together rail companies with seemingly similar maintenance needs, so that business knowledge is trapped, and then collaborate with an engineering solutions provider to incorporate prediction into it. However, to help rail companies graduate from their present level of IT usage and confidence to the level where such a DSS becomes acceptable, a TPI kind of a solution would prove to be the best bet. It would help these rail companies incorporate and derive benefits from predictive and planned maintenance capabilities. It can thus act as a bridge to graduate from the present level of low IT deployment to totally IT-enabled maintenance solutions.
Courtesy: TATA Consultancy Services (TCS)
|
Mail this article
Print this article
