Travel industry experts say that given the sometimes-mixed reputation of the country’s (China) ever-expanding air network, the train has becoming the travel mode-of-choice for a growing number of people, with its overnight ‘sleeper’ services gaining a particularly strong following.
There’s something special about spending a night on a train.
Maybe it’s the engine’s gentle rhythm rocking you to sleep, or the thought of falling asleep in one place and waking up in another.
From an official government news source concerning a government owned rail system. The article advocates a policy of what in the United States was once travel by Pullman.
China is roughly the size of the United States. China has far people than the United States. China’s population is heavily concentrated in its east. Congestion in China is of great concern.
Senior citizens and pleasure travelers are mentioned as a growing market for sleepers. Those with the time avoid the time in lines at airports and hotels.
America is certainly not China when it comes to passenger rail. Is such a market expansion strategy applicable to America’s government owned rail passenger service?
It’s an apples- and- oranges comparison given that even today,most Chinese citizens do not own automobiles and the country is far from having as developed an interstate highway system as the US(though they are roadbuilding at a furious rate)…
Are not the sleepers that Amtrak does operate often sold out? Is not Amtrak acutally short of sleepers? Should not there be at least one sleeper on the overnight Newport News - Boston train, actually an extension of the old Federal service? For Richmond - Boston and DC - Boston travel also?
It depends on what you mean by often. The sleepers are sold out, at least on some trains, during the high seasons, i.e. summer, Thanksgiving, and Christmas. But throughout the year probably not!
Although Amtrak’s monthly operating reports provide an easy way to calculate the average load factor for a train or service line (long distance trains), they don’t provide information to calculate easily the average load factor for the sleepers. It can, however, within reason be done train by train.
I don’t have the time or inclination to calculate the average sleeping car load factor for all the long distance trains. However, I pulled together some rough numbers for the Crescent, which had a 5.1 per cent increase in sleeping car revenues (higher per mile fares) in FY11 offset by a 3.2 per cent decline in sleeping car passengers.
There are four Crescents between the end points (two southbound and two northbound) during the schedule day, and there are 365 schedule days during a normal year. Based on my having ridden the train last year, each Crescent has two Viewliner sleeping cars. Maxed out each car can carry 32 passengers in 15 spaces. On average the Crescents offer 93,440 seats and 43,800 spaces during the year.
During FY11 the Crescent carried 32,616 sleeping car passengers for an average load factor of 34.9 per cent (seats) and 74.5 per cent (spaces). The former number assumes that every seat was occupied; the latter assumes that there was only one person per space. The actual average load factor would be in between these numbers. Again, if my experience is any indicator, most
While this thread is really about China (and perhaps we should return to that subject) I did look into sleeping accommodations on the Crescent over the next week.
Beginning tomorrow on the Crescent–Monday October 8–through next Sunday October 14: Saturday October 13 there is a roomette available. Sunday October 14 there is a bedroom available. As of now all other sleeping accommodations are sold out.
I would say your own assessment of sleeping accommodations is supported by my quick check of what is available on the Crescent. John
An assessment of Amtrak’s numbers doesn’t support a conclusion that Amtrak’s sleeping car space is at a premium, other than at certain periods, i.e, high season. A one time check of bookings is not a valid survey and, therefore, cannot be projected over the broader period, i.e. month, year, etc.
You can verify the numbers presented in my post on annual occupancy rates for the Crescent, which was in response to a question about Amtrak’s sleeping car occupancy rates.
As regular travelers on the Starlight and the Texas Eagle [LA to Dallas, r/t] we seem to need to book a bedroom at least 10 months in advance and even then be flexible as to the specific date of departure. I have no stats. but just experience. I know the Starlight is a premier train but I can assure the Eagle is not, yet the experience is the same.
I was only able to work up the stats for the total accommodations on the Crescent. I could do the same for the Texas Eagle, which saw a 6.2 per cent increase in sleeping car passengers during FY11, but I don’t have the time.
Part of the problem may lie in the fact that a Superliner sleeping car only has five bedrooms compared to 14 roomettes. Two of the bedrooms can be sold as a suite. In addition, there is a family room downstairs as well as an accessible bedroom. I suspect the accessible room is blocked except for mobility challenged passengers. Amtrak may hold the suite and family room to the last minute for a family, although I don’t know for sure.
Amtrak says that the roomettes and bedrooms are designed for two people, while the bedroom suite is designed for four people, and the family room is designed for two adults and two children. There are 28 seats in the roomettes vs.10 in the bedrooms excluding the family bedroom downstairs. The car can carry 42 adults, excluding those in the accessible room, and two children. Thus, the bedrooms could sell out (10 seats) and half of the roomette seats (14) could sell, for a total of 24 seats out of 44, excluding the accessible room, and the average load factor would be 55.5 per cent. If the train sold the six bedrooms and half the roomettes, the load factor would be 65 per cent of the available space.
As a matter of interest, why do you take the train from the west coast to Dallas on a seemingly regular basis. I have taken it to LA from Dallas and return on four or five occasions whilst on vacation or a
I just checked bedroom availability on the 9th of each month (beginning tomorrow, Monday) through August for the Texas Eagle. There are bedrooms available in February, March and May. All other months are sold out including June, July and August. The reservation system will not yet accept a reservation for September.
For most months roomettes are available but bedrooms are not. Since I used the 9th day each month the day of the week was different.
For the week of August 9 to 15 (11 months in advance) bedrooms are available 4 days but sold out on the other 3 days.
First, my wife and I are retired and our daughter and her family live in Frisco so it is just a nice, comfortable, hassle free way of travel. Secondly, as a rail geek, I really enjoy the process. This includes chatting with the crew, taking photographs for my Facebook site, and just watching the world go by! I know it’s crazy and we could fly for about half the price but who needs to be “strip searched” by the TSA!
No, but I have been patted down by shall we say very “friendly” TSA guys that seemed to be enjoying their jobs a great deal. I do fly over seas and if I’m in a hurry here in the states but flying is just no fun any longer. If I can I take the train
I don’t have the time or inclination to calculate the average sleeping car load factor for all the long distance trains. However, I pulled together some rough numbers for the Crescent, which had a 5.1 per cent increase in sleeping car revenues (higher per mile fares) in FY11 offset by a 3.2 per cent decline in sleeping car passengers.
for reasons that were not disclosed the Crescent only had and only booked one sleeper for some days in FY 2011. So capacity was artifically constrained and as expected number of passengers did decline.
There are four Crescents between the end points (two southbound and two northbound) during the schedule day, and there are 365 schedule days during a normal year. Based on my having ridden the train last year, each Crescent has two Viewliner sleeping cars. Maxed out each car can carry 32 passengers in 15 spaces. On average the Crescents offer 93,440 seats and 43,800 spaces during the year.During FY11 the Crescent carried 32,616 sleeping car passengers for an average load factor of 34.9 per cent (seats) and 74.5 per cent (spaces).
What kind of math is this? There is only one Crescent in each direction originating any day. Its times are similar to the 2 Florida trains. So 15 spaces times 2 cars times 2 trains ( 1 each way ) times 365 days = 21,900 spaces per year but as above only 1 car some days. That would mean that maxed out a total of 46720 passengers but that is not reasonable. When you traveled did you go solo ? If and this is only an if – which probably is diffferent on each LD route and day the number of passengers in a space may be 12 roomettes (18?) and 3 bedrooms ( 9? ) = 27 ??? This may mean on an average of practical yearly limits of what? 39420 passengers. That would give 32,616 / 39420 = 82% load factor ?? A standard deviati
This style of discourse where you take snippets of a post you disagree with and then interpose your views doesn’t add much to the discussion. Quote what you are responding to and then state your case.
I find it far easier to follow than quoting a solid block and then responding in a long post. This way, we can see each, discrete point and the corresponding response or comment directly after.
A look at the Crescent’s schedule shows that there are four trains enroute for all but about 45 minutes in a 24 hour day. They all sell space to and from every station except the northbound train would not sell space north of Alexandra. For example, at 2:30 p.m. Number 19 would be just shy of arriving in Newark, whilst Number 19 that had left the previous day would be between Tuscaloosa and Meridian. The northbound schedule is similar. They all sell space in the sleepers.
Normally there are eight Crescent sleepers on the road each day. They have 15 revenue spaces times 8 equals 120 spaces times 365 days equals 43,800 spaces or 87,600 seats assuming two people to a space. Push comes to shove a bedroom can accommodate more than two people.
I calculated an average load factor based on Amtrak information. Averages need to be looked at with a jaundice eye. One can drown in an average of six inches of water. Nevertheless, it suggests that Amtrak’s sleeper space is not bursting at the seems, except during peak periods as I noted.
Without an array, it is not possible to calculate a variance and standard deviation. It cannot be calculated from an average unless you have the array that was used to calculate the average. That means Amtrak would have to give us the passenger load for each day. It doesn’t.
On what days did the Crescent have only one sleeper? Making a claim without support does not carry a lot of weight.
The key question for a business is whether the cost of adding incremental space (coach or sleeper) would be offset by the incremental revenues. If it only reduces the amount of loss, it is not a good business decision. Without access to Amtrak’s assumptions, i.e. equipment cost, operating cost, turn aways, etc., it is impossible to know whether adding additional equipment would be a good business decision. I realize that making a good business decision is foreign to a company that loses more than $1.3 billion per ye
Page 33 diagrams 17 Roomettes available to the public and 6 Bedrooms, but that is after the improvements, currently I think they sell 16 Roomettes.Total rooms 22, Max Occupancy 44, yielding 44.260 million seat miles available or 22.130 million room miles.
Page 20 shows an average sleeper length of haul of 755 miles and 33,766 sleeper passengers = 25.493 million sleeper passenger miles in a year.
I typically assume 1.4 passengers per party, so 18.209 million room miles required. 18.209/22.130 = 82% occupancy of rooms could be plausible. The data to know for sure is not public. As to the degree that the Roomette is really a single passenger accommodation they did change the name.
See my paper on the US Intercity Market for a better breakdown of the ultimate marginal costs. I of course have a spreadsheet for that…
I overlooked the crew occupancy of the roomettes! In any case, we are not that far off in an estimate of high end occupancy numbers, i.e. 74.5% vs 82%. As you noted; as I noted, the occupancy rate today, which would be the basis for determining whether adding additional would be a good business decision, is a range that could be plausible.
Without access to Amtrak’s accounting, financial, and operational records, it is impossible to answer the key questions to know whether adding equipment (sleepers) is a good business decision. Determining that a losing business line (long distance trains) would lose less money but still lose money a good business decision does not make.
The solution to Amtrak’s long distance train issue is to drop the product line. That would be the decision of a competitive business. Unfortunately, Amtrak is not a competitive business. It is a ward of the state and oper
Sam you are mixing load with a load factor. You rode from WASH DC - New Orleans. You were one passenger on segments that you consider to be 2 passengers available.
Between NYP and new orleans there is only 30 spaces on each train available ( assumiing the normal 2 cars ). Since WASH - ATL is the night time portion those spaces are the preminum and dictate the number of sleeper spaces passengers want. That leaves the un needed spaces ATL - NOL empty. I’ll admit That lowers the revenue passengers miles down but not the revenue in same amount. Night time charges are greater per mile than the daY time charges check WASH - ATL vs NYP -NOL. vs miles
That being the case when the new sleepers on order are in service AMTRAK can place 3 - 4 sleepers NYP - ATL , then remove all but 1 and add them back on the northbound.that night. Removing multiple sleepers at ATL would economically justify the switching costs where as removing just one does not.
One sidebar that is not addressed is how many sleeper spaces are now taken by off duty crew membeers ? The addition of baggage - dorm cars will also increase the number of available spaces.
If you look at the 10 months of this FY you will find that all the sleeper route of the Star has about the same number of passengers as the Crescent. The extra sleeper on the Meteor & Lakeshore has that figure about 40 - 50% more and the Cardinal that operates with one sleeper 3 days a week but almost always sold out has a reduced number.
Sounds to me that more sleeper space much needed.
Next the $3.5 M cost of each sleeper vs added revenue from that space including initial coach fare needs to be computed but how far out fares should be checked is up to much debate. Maybe a comparsion of 6 months vs 1 month??.
It appears that passenger load factors are a bucket of worms. I want to approach this load factor question from another direction. I believe that neither of us has really looked at it closely so here goes.
Assumptions – 1 SINGLE LEVEL sleeping car NYP - ATL - NOL Octtober 16 fares (1st available ) Will use max and minimum passengers in a full car.
12 single occupancy roomettes and 3 two people in a bedroom.= 18 passengers. Space load factor = 100%. Passenger load factor what 100% or 50%
same space occupancy full but at maximum number of passengers = 36 passengers space load factor still 100%. Now what is the passenger load factor?. 100% ? 200% ?
2a. Full space but split NYP - ATL + ATL - NOL minimum passengers = min 36 max 72 & if space is full then space load factor is 100% but what are the passenger load factors ? 100% ? 200 %, 400% ?
18 passengers passage fare $223 to NOL Roomettee $553, Bedroom $814= $13092
Comparing either ex 2 with 4+ 6 or ex 3 with 5+7 we find most of fare is colllected NYP - ATL.
In conclusion taking the number of passenger and extrapolating to any kind of load factor is futile. Futher the fares are greatly biased higher on the north of Atlanta fares. Did not look at WASH but suspect same bias still there. Therefore the loads south of Atlanta do not matter as much as north.
Of all the eastern LD routes the Crescent north of ATL is probably the most convient for potential business travelers due to the departure and arrival times at ATL, WASH, BAL, &am