Fleet efficiency is very important: it relates to how much equipment is needed to provide a certain level of service along with spare cars in the need there are maintenance issues that require other equipment to be taken out of service. The more efficient a fleet is, the more service it can provide with a smaller number of locomotives, cars, etc. (and therefore lower capital and operating costs over the life of the fleet). The better maintained a fleet is, the less it breaks down, and the less of a shop margin is needed to substitute for malfunctioning cars.
The following data is from the Federal Transit Administration's National Transit Database. Transit operators across the country report a variety of information to the US DOT on an annual basis, and the most current information available is from 2018. This looks at just a small portion of the data that's reported, for operators classified as "commuter railroads" (plus the Alaska Railroad, which is classified separately). You can find the relevant data and the calculations used in this piece excerpted here.
The data is presented as-is, exactly as it was reported by the railroads. While the NTD effort does an overall great job of creating some sort of unified reporting and benchmarking mechanism for the nation's mass transit systems, there is always opportunities for error, either due to differences in calculations or methodologies, or agencies intentionally inflating/underballing the figures they report. I'll try to point out some of the glaring inconsistencies as we go along...for example, operators have to report the total number of mechanical failures that impact a scheduled trip to the US DOT...for 2018, the LIRR reported a total of 138 mechanical failures, yet in their monthly operating reports for 2018 included in the meeting notes for the MTA Board committee meetings, the LIRR reported a total of 423 primary failures (a mechanical issue that causes a delay), more than three times what the railroad reported to the US DOT...
Fleet efficiency
Train equipment is expensive, and it requires significant upfront investment. Once it is built, delivered, and placed into service, the marginal cost of running the train for 20 hours a day vs. 6 hours a day is not that much. To this end, the more train equipment is run per day, the more value we get for our upfront capital dollars. This is a big part of the reason behind "flattening the curve" and improving off-peak service to attract new off-peak service...off-peak service is cheap to add and can deliver big dividends, it almost always just requires using existing equipment and resources more efficiently. |
| LIRR's West Side Yard, filled to the brim with trains sitting around, doing nothing, instead of carrying passengers (Photo: Adam Fagen) |
According to data reported to the US DOT, the average LIRR train traveled an average of 64,572 miles in revenue service in 2018, which is good by American standards but still very poor when compared to modern regional rail systems. While that is well above the nationwide average of 40,997 miles in revenue service per year, the most fleet-efficient railroad in the country is Utah's UTA FrontRunner service, where their cars travel about 78,685 miles in revenue service per year.
When you divide that out by the total number of revenue service hours reported for the year, that means each LIRR train car is in service for just about 5.6 hours each day...in other words, they spend more than 18 hours per day (or 77% of their time) sitting around in yards doing nothing or running empty equipment trains, not carrying passengers.
| Railroad | Vehicles avail. for max svc. | Average revenue miles | Average daily hours in service |
|---|---|---|---|
| UTA FrontRunner | 69 | 78,685 | 6.54 |
| MTA Long Island Rail Road | 1,030 | 64,572 | 5.58 |
| MTA Metro-North Railroad | 1,165 | 58,196 | 5.13 |
| SMART | 14 | 54,774 | 8.60 |
| Caltrain | 134 | 53,749 | 4.40 |
| NICTD South Shore Line | 80 | 52,640 | 4.25 |
| Metrolink | 258 | 52,377 | 3.95 |
| MBTA Commuter Rail | 480 | 51,178 | 4.72 |
| SEPTA Regional Rail | 404 | 49,886 | 6.97 |
| NJTransit Rail | 1,297 | 47,943 | 3.99 |
| Trinity Railway Express | 34 | 47,854 | 5.94 |
| Tri-Rail | 80 | 45,092 | 4.26 |
| New Mexico Rail Runner Express | 31 | 43,504 | 3.18 |
| NCTD Coaster | 35 | 39,342 | 2.72 |
| Denver RTD | 66 | 38,836 | 2.94 |
| Metra | 1,204 | 36,275 | 3.30 |
| Shore Line East/Hartford Line | 47 | 32,817 | 2.02 |
| Altamont Corridor Express | 36 | 30,627 | 2.15 |
| MARC Train | 227 | 28,673 | 2.06 |
| Sounder Commuter Rail | 81 | 27,572 | 2.56 |
| Northstar Line | 24 | 24,992 | 1.78 |
| Virginia Railway Express | 120 | 20,136 | 1.79 |
| SunRail | 31 | 19,630 | 2.13 |
| Music City Star | 15 | 13,546 | 1.42 |
| Alaska Railroad | 96 | 12,021 | 1.22 |
The railroad that does the best job of keeping its trains on the move and not spending most of the day in yards is SMART in California, which started service in the summer of 2017...their cars are in service for an average of 8.60 hours per day. If LIRR were that efficient with its fleet, we could either increase service by about 35% per day, or reduce the number of vehicles in the fleet by more than 350 (the LIRR is spending $3.63 million per car for the M-9's, at that costs, this would translate to a $1.32 billion savings in capital costs and $170 million per year in maintenance costs).
I mentioned briefly in last week's piece about sizing a diesel fleet that the LIRR currently carries a large shop margin on their locomotive fleet, in part due to their poor reliability (and that I carried the same margin in the calculations for the new fleet by default). The LIRR needs roughly 30 DE/DM locomotives for weekday service, and they have a total of 45 on the roster, for a shop margin of 50%. On summer Fridays, the need rises to about 33 locomotives. The margin is a bit narrower for the C-3 fleet, with about 113 of the 134 cars needed for typical weekday service (rising to about 121 on Summer Fridays), a shop margin of about 19%. The railroad also needs a little over 900 of their 1,016 MU cars to mount peak service in electric territory, for a shop margin of about 12%.
 |
| The large new Morris Park locomotive shop, being built now to maintain LIRR's locomotive fleet (Photo: MTA LIRR) |
So how does the LIRR's maintenance spending and results stack up against railroads elsewhere in the United States?
Maintenance costs
When you look at what US railroads spend on maintenance on a per-vehicle basis, it probably won't surprise you to learn that the LIRR spends the most on maintenance per vehicle out of all railroads in the entire country... The LIRR spent an average of $469,954 per vehicle on maintenance in 2018, which is nearly double the next two railroads on the list, SunRail and LIRR's sibling operated by the same agency, Metro-North.US commuter railroads spent an average of $192,872 per vehicle on vehicle maintenance in 2018, making LIRR's costs 144% above the national average, and Metro-North's just 26% above average. The fact that LIRR's maintenance costs are nearly double costs per vehicle at Metro-North reinforces the tremendous efficiency gap between the two MTA railroads.
| Railroad | Vehicles avail. for max svc. | Maintenance labor cost per vehicle | Total maintenance cost per vehicle |
|---|---|---|---|
| MTA Long Island Rail Road | 1,030 | $359,039 | $469,954 |
| SunRail | 31 | n/a | $262,598 |
| MTA Metro-North Railroad | 1,165 | $180,190 | $245,971 |
| MBTA Commuter Rail | 480 | n/a | $227,558 |
| New Mexico Rail Runner Express | 31 | n/a | $222,224 |
| SMART | 14 | $184,331 | $214,480 |
| Caltrain | 134 | n/a | $211,811 |
| Shore Line East/Hartford Line | 47 | n/a | $198,504 |
| NJTransit Rail | 1,297 | $128,464 | $188,990 |
| NICTD South Shore Line | 80 | $131,571 | $180,023 |
| Trinity Railway Express | 34 | $374 | $169,449 |
| Metrolink | 258 | n/a | $167,954 |
| Tri-Rail | 80 | $1,150 | $165,174 |
| UTA FrontRunner | 69 | $73,270 | $158,819 |
| Sounder Commuter Rail | 81 | $2,178 | $151,605 |
| Metra | 1,204 | $105,538 | $143,161 |
| MARC Train | 227 | $4,474 | $141,176 |
| Northstar Line | 24 | $107,983 | $140,422 |
| SEPTA Regional Rail | 404 | $106,879 | $132,037 |
| NCTD Coaster | 35 | $1,714 | $125,260 |
| Denver RTD | 66 | n/a | $117,065 |
| Virginia Railway Express | 120 | n/a | $109,274 |
| Music City Star | 15 | n/a | $92,615 |
| Altamont Corridor Express | 36 | n/a | $92,329 |
| Alaska Railroad | 96 | $51,095 | $91,310 |
n/a = no labor breakout costs reported (usually means maintenance is contracted out)
There is obviously significant room for reducing LIRR maintenance costs...there should be a thing called economy of scale...the larger the railroad, the lower the per-unit maintenance costs should be since there is more efficiency with crews, facilities, etc. One of the nation's smallest railroads, like SunRail, having the high per-vehicle unit costs is somewhat understandable, but the nation's largest shouldn't be so far ahead of the pack. If the LIRR maintained their equipment at the same cost level as Metro-North, we would save over $231 million per year just in fleet maintenance costs. If the nation's largest railroad was also the nation's most efficient railroad on maintenance costs, we would save $389 million per year...
For context, all of the 2010 service cuts that the LIRR implemented during the last recession and will likely be looking at again with the ridership downturn following the COVID-19 pandemic (service cuts that included the decimation of off-peak service on the Port Washington Branch, the elimination of weekend service on the West Hempstead Branch and off-season weekend service between Ronkonkoma and Greenport, and many peak trains cut and combined) saved just $7 million per year.
Top dollar = best reliability?
Does the LIRR's extraordinary vehicle maintenance costs at least mean we have the most reliable fleet in the country? Not really. Dividing the total number of revenue miles by the total number of mechanical failures reported to the US DOT gives a number that's roughly equivalent to the Mean Distance Between Failure (MDBF) numbers that many railroads report, but not exactly. Under this metric, the LIRR does not have the best maintained or most reliable fleet in the country, they rank only seventh...behind Metro-North (which spends roughly half as much on maintenance as the LIRR does) and Virginia Railway Express (which is the fourth most efficient railroad in the country for maintenance vehicle costs...).When you relate the total maintenance costs per vehicle to the average distance a vehicle goes between breakdowns, the LIRR spends roughly $1.19 to keep a train moving for a mile without breaking down. Again, the LIRR is not matching its title as busiest railroad in the nation with the most cost-efficient railroad in the nation, ranking eleventh, behind Metro-North, VRE, an a number of small railroads.
| Railroad | Total Failures | MtDBF | Railroad | Cost per breakdown free mile | |
|---|---|---|---|---|---|
| Virginia Railway Express | 3 | 805,440 | Virginia Railway Express | $0.14 | |
| SMART | 1 | 766,833 | Alaska Railroad | $0.16 | |
| MTA Metro-North Railroad | 116 | 584,469 | SMART | $0.28 | |
| Alaska Railroad | 2 | 577,030 | Metra | $0.31 | |
| NICTD South Shore Line | 8 | 526,400 | NICTD South Shore Line | $0.34 | |
| Metra | 96 | 454,948 | MTA Metro-North Railroad | $0.42 | |
| MTA Long Island Rail Road | 168 | 395,886 | Altamont Corridor Express | $0.67 | |
| NJTransit Rail | 375 | 165,819 | Metrolink | $1.02 | |
| Metrolink | 82 | 164,797 | SEPTA Regional Rail | $1.08 | |
| Caltrain | 44 | 163,689 | NJTransit Rail | $1.14 | |
| Altamont Corridor Express | 8 | 137,822 | MTA Long Island Rail Road | $1.19 | |
| SEPTA Regional Rail | 165 | 122,145 | Caltrain | $1.29 | |
| Trinity Railway Express | 14 | 116,218 | Sounder Commuter Rail | $1.36 | |
| Sounder Commuter Rail | 20 | 111,667 | Trinity Railway Express | $1.46 | |
| MBTA Commuter Rail | 273 | 89,983 | Music City Star | $1.82 | |
| SunRail | 8 | 76,068 | MARC Train | $1.93 | |
| MARC Train | 89 | 73,132 | MBTA Commuter Rail | $2.53 | |
| NM Rail Runner Express | 21 | 64,220 | UTA FrontRunner | $2.69 | |
| UTA FrontRunner | 92 | 59,013 | Tri-Rail | $3.25 | |
| Tri-Rail | 71 | 50,808 | SunRail | $3.45 | |
| Music City Star | 4 | 50,799 | NM Rail Runner Express | $3.46 | |
| NCTD Coaster | 39 | 35,307 | NCTD Coaster | $3.55 | |
| Shore Line East/Hartford Line | 61 | 25,285 | Shore Line East/Hartford Line | $7.85 | |
| Denver RTD | 373 | 6,872 | Denver RTD | $17.04 | |
| Northstar Line | 93 | 6,450 | Northstar Line | $21.77 |
And keep in mind that the number of mechanical failures the LIRR reported to the MTA Board was significantly higher than the number they reported to the US DOT (the figures used here), so in reality, they rank significantly lower on the list.
This is something that the US commuter railroad industry tends to struggle with in general... When you relate maintenance spending to reliability, there is very little correlation across the country. Spending more on maintenance does not really tend to yield much in terms of improved fleet reliability, suggesting that this is largely a practice and efficiency issue from system to system.
It likely comes at no surprise to anyone reading that the LIRR spends the most on vehicle maintenance out of any railroad in the United States BY FAR, yet riders get results that are mediocre, at best... The MTA has a number of years of great financial difficulty ahead of it as a result of the COVID-19 pandemic, and they will undoubtedly be looking for handouts from the federal government, service cuts to save money, and, of course, additional fare increases to plug the gaping holes in their budget. As the numbers show, LIRR maintenance value is pretty poor, and as part of the recovery effort the railroad must be challenged to either dramatically reign in these crazy fleet maintenance costs, or dramatically improve fleet reliability and results, so that we can keep our existing cars in service for many more years and get by with much narrower shop margins (reducing the need to purchase extra equipment as spares). The amount of money the LIRR could save with even tiny economizations in fleet maintenance can easily dwarf the amount they can save in service cuts. The recovery from the pandemic must include management mounting significant challenges to the status quo, not just cutting service, hiking fares, or looking for more subsidy (so the farepayers and taxpayers bear all of the impacts) and carrying on with business as usual.
No comments:
Post a Comment