«COMFORTABLE SARDINES: THE BALANCE BETWEEN COMFORT AND CAPACITY Report submitted to: A A Leyland Southampton Railway Systems Research School of Civil ...»
COMFORTABLE SARDINES: THE BALANCE BETWEEN COMFORT
Report submitted to:
A A Leyland
Southampton Railway Systems Research
School of Civil Engineering and the Environment
University of Southampton
Report prepared by:
H V C Howarth and M J Griffin
Human Factors Research Unit
Institute of Sound and Vibration Research
University of Southampton Southampton C Childs and T Fujiyama Accessibility Research Group Centre for Transport Studies Civil, Environmental and Geomatic Engineering Department University College London S G Hodder Environmental Ergonomics Laboratory Loughborough Design School Loughborough University 12 April 2011 Contents 1. INTRODUCTION
2. FACTORS AFFECTING THE COMFORT OF RAIL PASSENGERSWHEN TRAVELLING AT HIGH DENSITY
2.1 Passenger movement into and around a crowded train
2.1.2 Experimental research
2.1.4 Conclusions and recommendations for future work
2.2 Thermal comfort in a crowded train
2.2.2 Experimental work
2.2.4 Conclusions and recommendations for future work
2.3 Comfort and stability when standing and walking in moving environments
2.3.2 Previous studies
2.3.3 Existing models
2.3.4 Conclusions and recommendations for future work
2.4 High density seating: perches
2.4.2 Conclusions and recommendations for future work
3. STUDIES TO DEVELOP THE RAIL CARRIAGE OF THE FUTURE............ 10
3.1 Passenger movement
3.2 Thermal comfort
3.3 The value of comfort
3.4 Project partners for future proposals
APPENDIX MINUTES OF RAIL RESEARCH SYMPOSIUM OF 9THDECEMBER 2010
Summary This report describes the findings of a feasibility study conducted to review factors influencing the comfort of rail passengers when passenger density is high, including situations where passengers stand during the journey. The state of knowledge of passenger movement and thermal comfort in crowded trains, comfort when standing and walking in a moving environment, and high density seating is summarized.
Research is proposed to assist optimisation of the trade-off between passenger density and passenger comfort on trains. The proposed studies take into account comments and suggestions provided by rail industry representatives.
The study was one of a portfolio of feasibility studies conducted under the EPSRCfunded RRUK Feasibility Account ‘Factor 20 – reducing CO2 emissions from inland transport by a major modal shift to rail’ (EP/H024743/1).
1. INTRODUCTION Over the past 10 years, rail passenger numbers have risen by about 40% and the industry is expecting demand to double over the next few decades. Peak-time timetables are at maximum capacity and it is difficult to lengthen trains. Levels of overcrowding are likely to increase with a continuing rise in the demand for rail travel.
Increasing passenger density on a train will reduce the CO 2 emission per passenger.
However, any reduction in passenger comfort associated with increased passenger density is expected to encourage travel by car, which would increase CO 2 emission per passenger. Minimisation of the CO 2 emission per passenger requires an understanding of the trade-off between passenger density and passenger comfort.
This report describes the findings of a feasibility study conducted to review factors influencing the comfort of rail passengers when passenger density is high, including situations where passengers stand during the journey. The state of knowledge on passenger movement and thermal comfort in crowded trains, comfort when standing and walking in a moving environment, and high density seating is summarized.
The study involved three University partners who brought complementary expertise to the study: the Human Factors Research Unit, in the Institute of Sound and Vibration Research, at the University of Southampton; the Pedestrian Accessibility and Movement Environment Laboratory at University College London; and the Environmental Ergonomics Research Centre at Loughborough University.
A symposium was organised with representatives of the railway industry to discuss ideas and proposals for future investigation (see Appendix). Research is proposed to assist the optimisation of the trade-offs between passenger density and passenger comfort on trains. The proposed studies take into account comments and suggestions of rail industry representatives.
2. FACTORS AFFECTING THE COMFORT OF RAIL PASSENGERS WHEN
TRAVELLING AT HIGH DENSITY
2.1 Passenger movement into and around a crowded train 2.1.1 Standards The statutory framework for the standards that refer to train carriage access for people with disabilities (HMSO, 1998; Department for Transport and Transport Scotland, 2008) is the Disability Discrimination Act 1995; 2005 (Great Britain Parliament, 1995; Great Britain Parliament, 2005). These standards specify that doorways need to be greater than 850 mm wide and that the maximum horizontal and vertical distance between the platform and the carriage floor should be 50 mm 1 (horizontal) and 75 mm (vertical) otherwise a lift or ramp is required. In addition, 10% of the available seating should be priority seating for the elderly and disabled, with at least two wheelchair spaces (but depending on the number of carriages).
Emergency evacuation of train carriages is covered by the European Commission Directive 96/48/EC (The Commission of the European Communities, 2002). This states that the door dimensions shall be at least 700 mm × 550 mm and that they allow the complete evacuation of passengers in normal operation within three minutes. No passenger seat will be more than 16m from an emergency exit and there should be at least two emergency exits for each carriage. The Strategic Rail Authority state (UK Department for Transport, 2005) that automatic doors should take 3 seconds to open and remain open for at least 6 seconds, although they do not explain the rationale behind these requirements.
The ‘Level of Service’ concept was developed for assessing vehicle flows and subsequently adapted by Fruin (1971) to categorise space used by pedestrians. This concept categorises a space according to the number of people moving within it, assigning a functional classification from low density ‘A’ to high density ‘F’, where ‘F’ has been described as ‘crush’ conditions. The Highway Capacity Manual (TRB,
2000) has incorporated these categories into chapter 13. However there is disagreement over what each category means in practice. For example, higher individual maximum speeds and space occur in category ‘A’ whilst the highest flow rate occurs in category ‘E’. There is currently no equivalent system directly applicable to either travelling on, or boarding and alighting from, trains.
2.1.2 Experimental research Experimental work has focussed on evacuation times or steady state movement of pedestrians on level surfaces, through bottlenecks, up/down stairs, or considered the dwell time of trains in stations. The problem with bottleneck studies, (for example, Winkens et al., 2008) is that the free flow of people once they are past the bottleneck is not applicable to train boarding. In addition, evacuation studies (Oswald et al.,
2008) do not include the movements of people onto the train. Emergency evacuation times are not comparable to everyday flow rates as the number of people boarding is low in evacuation conditions but not in everyday use, and the psychological forces driving passenger movements are significantly different. References in Section 2.1.3 below on ‘Models’ include elements of experimental justification. The importance of capacity is unknown – how many people can get into train carriages given unlimited time (acceptable comfort levels for passengers) or the capacity given different internal layouts. A UCL project on train carriages for the Department for Transport 2 examined boarding and alighting with different door/vestibule dimensions and indicated that it would be difficult to get 50 passengers through the doors in the requested 27 seconds (Fujiyama et al (2008)). This finding fits with the summary of board/alight times discussed in Wiggenraad (2001) of approximately 1 s per person measured in train stations. Lee et al. (2007) observed travellers at Dutch train platforms and suggested that vehicle design and crowding non-linearly affect board and alight times. Evans and Wener (2007) found that seat density (the ratio of the number of people to seats in the immediate area of the passenger) has a more important influence on passenger stress than car density (the ratio of the number of people on a passenger car to the total number of seats). Cox et al. (2006) suggested that high-density and crowding are not necessarily synonymous, suggesting that a high density might be achieved without perceptions of crowding.
Most pedestrian/passenger models try to address how pedestrians evacuate buildings or ships in case of emergency and fire, in particular focussing on capacity issues at bottlenecks such as doorways. There are continuous and discrete methods.
Continuous models calculate position based on so called ‘social forces’, parameters that reflect interpersonal psychological relations that attract or repel the pedestrian from their surroundings (Helbing and Molnár, 1995). Discrete methods apply cellular automaton techniques with rules as to how pedestrians move from one cell to another in arrays created to represent the pedestrian space (Fukui and Ishibashi, 1999).
These models either consider pedestrians as steady state flow (for example along corridors) encountering obstacles (for example bottlenecks such as doorways that affect flow rates), or in terms of evacuation. Consequently they are not interested in where the pedestrian ends up. The ‘social forces’ each passenger is subjected to may be different when boarding a carriage, travelling on the carriage, and when alighting. These states are not sufficiently understood to be incorporated into these models. In addition, the train evacuation models have not included people in wheelchairs (Capote et al., 2008).
One cellular-automata-based model considers boarding and alighting the Beijing metro (Zhang et al., 2008). The group size boarding/alighting was varied and compared to four different platforms over three stations. The trains and platforms were the same; therefore carriage layout, doorway width, and height from the platform were not considered. Models may appear to have the benefit of creating optimised results prior to a solution being built, but it is important to remember that all 3 the models are based on assumptions and simplifications. Rogsch et al. (2005) reported the variations in evacuation times calculated from a number of simulation programs set with simple geometries.
2.1.4 Conclusions and recommendations for future work ‘Level of Service’ appears to be a useful concept for comparing pedestrian movements in different locations, but it is not directly applicable to train travel, where people board, stay, and then alight. It would be useful to expand this concept, to incorporate these phases and, importantly, include passenger comfort. The limitations of existing pedestrian models are that they are validated by macroscopic indices (for example, it took XXX seconds to evacuate the building in a fire drill and a simulation shows a similar result). In addition, all the obstacles in the existing simulations are continuous. It is likely that people's reactions change according to the shape and other properties of obstacles. Passenger comfort is not well understood in terms of physical proximity when boarding, travelling on, or alighting train carriages and the perception of comfort may be expected to vary with each of these actions, travel time, and the carriage layout. The interactions between people and between a person and a facility are not understood. Such interactions can be analysed in the Pedestrian Accessibility and Movement Environment Laboratory of the University College London. A pedestrian simulation model can be developed to include functions for stable pedestrians and their reactions to external environmental stimuli as well as their interactions with other passengers and the characteristics of the carriage.
2.2 Thermal comfort in a crowded train 2.2.1 Standards British Standards BS EN 13129 Parts 1 and 2 (2002) and BS EN 14750 Parts 1 and 2 (2006) provide guidance on the ‘bench marking’ of train climate control. Tests are conducted on stationary unoccupied carriages and provide guidelines for internal temperature requirements for a range of external temperatures (-40°C to 40°C). Air flow rates are also considered.
Other standards for the evaluation of train environments are based on the responses of humans rather than the carriage (ISO 7730, 2005; ISO 7933, 2004). These standards provide guidance on how the occupant will react to a given thermal environment in the carriage.
The standards consider mainline, urban, and suburban railway carriages. The standards are applicable to both passengers and railway staff, with the exception of cooking areas. Additional standards for driver cabins are also available.
18.104.22.168 Limitations The current applicable standards do not adequately consider occupied carriages.
The published data on the thermal responses of people in densely occupied spaces provides only limited support for current standards or the development of new standards.
2.2.2 Experimental work Few studies have specifically investigated the thermal comfort of train passengers.
The focus of most published work has been on how to quantify the thermal environment to optimise thermal comfort (Berlitz and Matschke, 2002; Chow, 2002;