sustainability

Article Influence of Evacuation Walkway Design Parameters on Passenger Evacuation Time along Elevated Rail Transit Lines Using a Multi-Agent Simulation

Zihua Pan 1,2,* , Qingchao Wei 1, Olav Torp 2 and Albert Lau 2

1 School of Civil Engineering, Beijing Jiaotong University, Beijing 100044, China; [email protected] 2 Department of Civil and Environmental Engineering, Norwegian University of Science and Technology, 7491 Trondheim, Norway; [email protected] (O.T.); [email protected] (A.L.) * Correspondence: [email protected]; Tel.: +86-152-0134-3069



Received: 30 September 2019; Accepted: 29 October 2019; Published: 31 October 2019


Abstract: Passenger evacuation on elevated railway lines has always been an important issue for elevated rail transit safety management, because it is challenging to evacuate passengers efficiently in the event of man-made calamities and natural disasters. Therefore, an evacuation walkway has been designed as a primary solution to assist passenger evacuation during an emergency on elevated rail transit lines. However, investigations on how evacuation walkway designs influence passenger evacuation time are still limited. This study established two evacuation scenarios of interval evacuation on elevated rail transit lines and put forward a new evacuation time measurement method, based on the concept of ‘evacuation time for passengers leaving the evacuation walkway risk zone’. Then, the evacuation time for 90 combinations of entrance widths and walkway widths was simulated by a multi-agent evacuation simulator, Pathfinder, considering 1032 passengers being evacuated both unidirectionally and bidirectionally. The results show that the entrance width and walkway width have a combined effect on passenger evacuation time. An increase in the walkway width from 0.7 m to 1.5 m may potentially reduce the evacuation time by 54.5% in unidirectional evacuation, and 35.2% in bidirectional evacuation. An increase in the entrance width results in a noticeable evacuation time fluctuation when the walkway width is 0.7 and 0.8 m for both evacuation scenarios, while in a bidirectional evacuation, a noticeable fluctuation also can be observed when the walkway width is within the range of 1.4–1.5 m. According to the study, a potentially good design parameter combination for a newly built evacuation walkway is 1.3 m and 1.4 m for the walkway width and entrance width, respectively. The findings from this study may provide a useful reference in the optimization of the design of evacuation facilities and improvement of passenger evacuation safety in rail transit systems.

Keywords: passenger safety; Maglev transit; evacuation time; multi-agent simulation; evacuation walkway; parametric design

1. Introduction Due to the increase in man-made accidents and natural disasters, passenger evacuation in a rail transit system has became a more prominent research issue for emergency response planners, transport engineers and policymakers [1–6]. It is challenging to evacuate a mass volume of passengers in a well-organised and highly-efficient way in a rail transit system because of the uncertainty of accidents and complex nature of individual characteristics [7–9]. Passenger evacuation in rail transit systems can be hard to predict, because the types and locations of accidents that occur in rail transit are unpredictable [10]. In many cases, passengers are required to evacuate quickly in a difficult-walking

Sustainability 2019, 11, 6049; doi:10.3390/su11216049 www.mdpi.com/journal/sustainability Sustainability 2019, 11, x 6049 FOR PEER REVIEW 22 of of 18 17 walking environment, e.g., through a tunnel with low visibility or a viaduct with limited walking spaceenvironment, [11,12]. The e.g., tendency through a for tunnel passengers with low to visibilitybe caught or in a viaductdanger, withe.g., falling limited over walking or panicking space [11 in,12 a]. congestedThe tendency and forunfamiliar passengers place, to beis caughthigh. Furthermore, in danger, e.g., distance falling to over safety or is panicking an uncertain in a congestedfactor for passengers.and unfamiliar While place, the priority is high. is Furthermore, to evacuate passenger distances to to safety the nearest is an uncertainstation in case factor of foran emergency, passengers. itWhile remains the prioritya possibility is to evacuatethat a train passengers will be forced to the to nearest stop at station interval in sections case of an due emergency, to a train itfire, remains train collision,a possibility power that failure, a train etc. will [13,14]. be forced In this to stopcase, at people interval might sections need dueto walk to a for train a long fire, distance train collision, to get awaypower from failure, the etc. dangerous [13,14]. Inaccident this case, site people before might they re needach tothe walk safety for exits. a long Therefore, distance to it getis essential away from to providethe dangerous a user-friendly accident and site beforehighly-efficient they reach evacuation the safety facility exits. Therefore, for passengers’ it is essential safety in to rail provide transit a systemsuser-friendly [15]. and highly-efficient evacuation facility for passengers’ safety in rail transit systems [15]. The evacuation walkway for Elevated Electrom Electromagneticagnetic Suspension (EMS) Maglev transit may be one typical evacuation facility for rail transit systems. The EMS maglev track structure is known for having less space for evacuation in comparison to a conventional ballast or ballastless rail track structure (see Figure 11).). TheThe solutionsolution toto thatthat isis thereforetherefore anan extraextra elevatedelevated evacuationevacuation walkway,walkway, nextnext to the rail transit systems. It It is is worth worth noting noting that that an an elevated elevated walkway walkway that that is is high high from from the ground might pose a problematic evacuation condition for passengers who are acrophobicacrophobic [[16].16].

Figure 1.1. AA track track structure structure comparison comparison of a metroof a systemmetro andsystem an Elevatedand an ElectromagneticElevated Electromagnetic Suspension Suspension(EMS) Maglev (EMS) system: Maglev (a) system: a typical (a rail-track) a typical structurerail-track ofstructure a traditional of a traditional metro system metro [17 system]; (b) EMS[17]; (Maglevb) EMS trackMaglev structure track structure of the Beijing of the Maglev Beijing Express.Maglev Express.

Some research hashas shownshown that that an an appropriately appropriately designed designed evacuation evacuation walkway walkway could could improve improve the theevacuation evacuation conditions conditions in rail in transit rail transit constructions. constructi Forons. instance, For instance, Habicht Habicht et al. [18 ]et proposed al. [18] proposed a definition, a definition,together with together a method, with to calculatea method, the to eff ectivecalculate width the reductions effective underwidth variousreductions conditions, under basedvarious on conditions,pedestrian distributionbased on pedestrian observations distribution in a pedestrian observat tunnel.ions in They a pedestrian found that tunnel. the actual They efoundffective that width the actualof the passagewayeffective width is generally of the widerpassageway than that is assumedgenerally in wider some designthan that practices. assumed Lundstrom in some et design al. [19] practices.discussed theLundstrom relationship et al. between [19] discussed a raised walkwaythe relationship width and between people’s a raised evacuation walkway behaviour width in and rail people’stunnels using evacuation a model behaviour of a tunnel in walkway.rail tunnels They using found a model that aof walkway a tunnel widthwalkway. of around They found 1 m leads that toa walkwaya higher moving width of speed, around and 1 there m leads is a linearto a higher relationship moving between speed, the and dynamic there is flow a linear of the relationship crowd and betweenthe walkway the dynamic width. Moreover, flow of the the crowd evacuation and the walkway walkway width width. consideration Moreover, the is a evacuation mandatory walkway aspect of widthevacuation consideration regulations is a in mandatory many countries. aspect of For evacuation example, theregulations minimum in walkwaymany countries. width inFor a example, tunnel of 20 be km at least should 0.75 be m, at while least 1.1that m for according tunnels >20to EU km regulations should be at [20 least]. According 1.1 m according to the designto EU re codegulations of the [20]. Metro According of China to and the the design design code code of the for MetroMedium of andChina Low-Speed and the Maglevdesign Transitcode for in China,Medium the and minimum Low-Speed evacuation Maglev walkway Transit width in China, should the be minimum0.7 m, which evacuation is the same walkway as the recently width builtshould Copenhagen be 0.7 m, Metrowhich [ 21is –th23e]. same as the recently built CopenhagenThe research Metro mentioned [21–23]. above is useful for designing the evacuation walkway width, while the connectingThe research parts betweenmentioned the above train is and useful the for evacuation designing walkway the evacuation may result walkway in a bottleneckwidth, while eff theect connectingduring the evacuationparts between process the train [23]. and For the example, evacuati Fridolfon walkway et al. [24 may] found result that in train a bottleneck exit flows effect and duringwalkway the flows evacuation are likely process to be constrained[23]. For example, due to mergingFridolf et conditions al. [24] found in the that evacuation train exit process, flows and walkwaythat merging flows increases are likely the to required be constrained time for due evacuees to merging to reach conditions a safe place.in the Arturoevacuation Cuesta process, et al. and [25] thatalso merging confirmed increases the negative the required effect of time the occurrencefor evacuees of to merging reach a onsafe evacuation place. Arturo time Cuesta through et al. a field[25] also confirmed the negative effect of the occurrence of merging on evacuation time through a field

Sustainability 2019, 11, 6049 3 of 17

Sustainability 2019, 11, x FOR PEER REVIEW 3 of 18 experiment in a mock rail car, with a single exit towards a lateral corridor. According to their data, the railexperiment car exit flow in a and mock the rail walkway car, with flow a single decreases exit towards when merging a lateral occurs. corridor. According to their data, theAccording rail car exit to flow the above-mentionedand the walkway flow studies, decreases it may when be the merging case that occurs. the walkway width, along with the entranceAccording width, to ofthe an above-mentioned evacuation walkway studies, may it togethermay be the aff ectcase the that passenger the walkway evacutation width, along process; however,with the to entrance date, only width, limited of an research evacuation can be walkway found that may discusses together thisaffect combined the passenger effect onevacutation evacuation process; however, to date, only limited research can be found that discusses this combined effect on time in urban rail systems. As an example, the entrance width of an evacuation walkway is usually set evacuation time in urban rail systems. As an example, the entrance width of an evacuation walkway as a constant or two to three different values, and the evacuation direction is not a consideration [24,25]. is usually set as a constant or two to three different values, and the evacuation direction is not a For countries like China, the minimum entrance width of evacuation walkway is 600 mm, and for consideration [24,25]. For countries like China, the minimum entrance width of evacuation walkway Japan the minimum entrance width of evacuation walkway is 660 mm or 800 mm for passengers with is 600 mm, and for Japan the minimum entrance width of evacuation walkway is 660 mm or 800 mm a wheelchairfor passengers [21, 26with] a wheelchair [21,26] Therefore,Therefore, the the objective objective of of this this studystudy isis toto findfind the combined ef efectffect of of the the entrance entrance width width and and walkwaywalkway width width of of an an evacuation evacuation walkway walkway on on evacuation evacuation time time and and then then to to put put forwardforward aa suggestionsuggestion as to aas potentially to a potentially better better design design combination, combination, which which may may result result in a shorterin a shorter evacuation evacuation time. time. The The study flowstudy and flow its corresponding and its corresponding contribution contribution are as follows:are as follows: (i) Establish passenger evacuation scenarios in elevated rail transit lines and conduct a field survey (i) Establish passenger evacuation scenarios in elevated rail transit lines and conduct a field on elevated Electromagnetic Suspension (EMS) transit lines in China. This aids in the collection of survey on elevated Electromagnetic Suspension (EMS) transit lines in China. This aids in the necessary information on passengers for the simulations in the second step. collection of necessary information on passengers for the simulations in the second step. (ii)(ii) Use Use a multi-agenta multi-agent simulator, simulator, Pathfinder, Pathfinder, toto calculatecalculate the the evacuation evacuation time time for for 90 90 combinations combinations of entranceof entrance widths widths and and the the walkway walkway widthswidths ofof an evacuationevacuation walkway. walkway. This This aids aids in in the the study study of ofthe the effecteffect of of the the design design combinations combinations parametrically. parametrically. (iii)(iii) Analyse Analyse the the combined combined e effectffect ofof thethe twotwo designdesign parameters on on evacuation evacuation time time using using a new a new evacuationevacuation time time measurement. measurement. ThisThis aidsaids inin suggestingsuggesting potentially potentially good good combinations combinations of ofentrance entrance widthswidths and and walkway walkway widths widths for for the the design design ofof anan evacuationevacuation walkway walkway..

2. Methodology2. Methodology TheThe methodology methodology of of this this study study consistsconsists ofof 22 parts.parts. First, First, typical typical evacuation evacuation scenarios scenarios were were surveyed,surveyed, then then passenger passenger evacuation evacuation timetime waswas si simulatedmulated by by a a multi-agent multi-agent evacuating evacuating simulator, simulator, Pathfinder.Pathfinder. The The research research processprocess ofof thisthis study is shown in in Figure Figure 2.2. Step Step (iii) (iii) of of the the study study flow, flow, mentionedmentioned in in the the introduction, introduction, is is addressed addressed inin thethe resultsresults and the discussion. discussion.

Figure 2. Research process. Figure 2. Research process. 2.1. Evacuation Scenario Survey

The purpose of the evacuation scenario survey was to obtain basic information to support the building of an evacuation simulation model, e.g., typical evacuation routes selected by the passengers and different evacuation walkway design features.

Sustainability 2019, 11, x FOR PEER REVIEW 4 of 18

Sustainability2.1. Evacuation2019, 11, 6049Scenario Survey 4 of 17 The purpose of the evacuation scenario survey was to obtain basic information to support the buildingThe survey of an isevacuation divided simulation into three model, steps: e.g., First, typical an evacuation interview routes of Maglev selected transit by the operators passengers and constructionand different companies evacuation in Changsha walkway anddesign Beijing, features. respectively. Information on, for example, emergency types, frequencyThe survey of emergencies, is divided into and three passenger steps: evacuationFirst, an interview patterns, of is discussedMaglev transit and collected.operators Second,and the designconstruction of the companies evacuation in walkway Changsha is and surveyed Beijing, by respectively. visiting the actualInformation evacuation on, for routes example, on the emergency types, frequency of emergencies, and passenger evacuation patterns, is discussed and Maglev transit lines in Shanghai, Changsha and Beijing. Finally, a questionnaire survey was conducted collected. Second, the design of the evacuation walkway is surveyed by visiting the actual evacuation to obtain the basic features of passengers using Maglev transit, e.g., age, gender, level of education and routes on the Maglev transit lines in Shanghai, Changsha and Beijing. Finally, a questionnaire survey simplewas self-estimated conducted to obtain choice the of basic evacuation features routes.of passen Agers total using of 400 Maglev questionnaires transit, e.g., age, were gender, distributed level at 5 Maglevof education transit and stations simple in self-estimated Changsha and choice Shanghai, of evacuation and 338 routes. valid A ones total wereof 400 retrieved. questionnaires were distributedFrom the evacuationat 5 Maglev transit scenario stations survey, in Changs the followingha and Shanghai, features and of 338 an valid interval ones evacuationwere retrieved. on an elevated MaglevFrom the transit evacuation line are scenario considered survey, in the the followin numericalg features evacuation of an simulationinterval evacuation model. on an elevatedA train isMaglev forced transit to stop line at are an considered elevated interval. in the numerical The passengers evacuation must simulation alight model. from the carriages, enter theA evacuation train is forced walkway to stop andat an evacuate elevated interval along the. The walkway, passengers until must they alight reach from the the safety carriages, exits at the stations.enter the evacuation The evacuation walkway starts and from evacuate the along middle the point walkway, of the until interval, they reach which the issafety at the exits longest at walkingthe stations. distance The from evacuation both of the starts two nearestfrom the stations. middle Twopoint scenarios of the interval, are considered which is inat thethe simulation,longest a unidirectionalwalking distance and bidirectional from both of scenario, the two in nearest which thestations. passengers Two scenarios can evacuate are considered with only onein the safety simulation, a unidirectional and bidirectional scenario, in which the passengers can evacuate with exit (the left one) or two safety exits, respectively. It is assumed that no working staff organized and only one safety exit (the left one) or two safety exits, respectively. It is assumed that no working staff coordinated the evacuation. The evacuation scenario is shown in Figure3. organized and coordinated the evacuation. The evacuation scenario is shown in Figure 3.

FigureFigure 3. Schematic 3. Schematic drawing drawing of of an an interval interval evacuation evacuation scenario(s)scenario(s) on on an an elevated elevated Maglev Maglev transit transit line. line.

2.2. Multi-Agent2.2. Multi-Agent Evacuation Evacuation Simulation Simulation

2.2.1.2.2.1. Model Model Basis Basis In thisIn this study, study, an agent-basedan agent-based evacuation evacuation simulator, simulator, Pathfinder,Pathfinder, was was used used to to simulate simulate the the process process of passengerof passenger evacuation evacuation on on an an elevated elevated rail rail transit transit line.line. It It is is known known to to be be able able to tosimulate simulate the the evacuationevacuation time time for for a mass a mass passenger passenger volume volume without without having to to set set up up a arisky risky and and expensive expensive field field experiment,experiment, e.g., e.g., construct construct an actualan actual evacuation evacuation walkway walkway with with diff differenterent designs designs and and avoid avoid safety safety risks risks in a crowded situation. in a crowded situation. The steering mode of Pathfinder was chosen to simulate passenger movement. The steering The steering mode of Pathfinder was chosen to simulate passenger movement. The steering mode mode is an agent-based model which is following Reynold’s steering behaviour model [27] and is anrefined agent-based by Amor model et whichal. [28]. is The following passengers Reynold’s were set steering as an behaviour autonomous model agent [27 with] and different refined by Amorcharacteristics et al. [28]. The such passengers as age, moving were set speed as an autonomousand pre-movement agent withtime di[29].fferent In characteristicsthe steering mode, such as age, movingPathfinder speed uses anda combination pre-movement of steering time [ 29and]. Inco thellision steering handling mode, mechanisms Pathfinder to usescontrol a combination how the of steering and collision handling mechanisms to control how the passengers achieve their goal in an emergency situation. These mechanisms allow the passengers to move along their path, deviate from the obstructions (interact with the environment and the other passengers) and still head in the correct direction toward their goal, the safety exits [30]. The algorithms and decision-making process when passengers follow the steering mode have been summarised in Figure4, and the calculation process can be found in the technical reference of Pathfinder [31]. Sustainability 2019, 11, x FOR PEER REVIEW 5 of 18

passengers achieve their goal in an emergency situation. These mechanisms allow the passengers to moveSustainability along 2019 their, 11 ,path, x FOR PEERdeviate REVIEW from the obstructions (interact with the environment and 5the of other18 passengers) and still head in the correct direction toward their goal, the safety exits [30]. The algorithmspassengers and achieve decision-making their goal in an process emergency when situ ation.passengers These mechanismsfollow the steering allow the mode passengers have tobeen summarisedmove along intheir Figure path, 4,deviate and thefrom calculation the obstructions process (interact can be with found the environmentin the technical and thereference other of Pathfinderpassengers) [31]. and still head in the correct direction toward their goal, the safety exits [30]. The algorithms and decision-making process when passengers follow the steering mode have been Sustainabilitysummarised2019, 11in, 6049Figure 4, and the calculation process can be found in the technical reference 5of of 17 Pathfinder [31].

Figure 4. The decision-making process, when passengers evacuate following the steering mode in FigurePathfinder.Figure 4. 4.The The decision-makingdecision-making process, process, when when passenger passengerss evacuate evacuate following following the steering the steering mode mode in in Pathfinder.Pathfinder. In Pathfinder, the construction environment of the elevated rail transit lines can be represented withIn aIn Pathfinder,3D Pathfinder, geometry the the mode, construction construction i.e., evacuation environment environment walkway, ofof thethe station elevated platform rail rail transit transit and lines lines stairs can can atbe bedifferent represented represented layers with[31].with a A 3D a navigation3D geometry geometry mode,mesh mode, is i.e., definedi.e., evacuation evacuation as a walkway,continuo walkway,us station station2D triangulated platform platform and and surface stairs stairs at within at di ffdifferenterent a 3D layers layersgeomtry [31 ]. Amodel navigation[31]. A(see navigation Figure mesh 5).is mesh Passengers’ defined is defined as a motion continuous as a takescontinuo 2Dplaceus triangulated on2D this triangulated navigation surface surface mesh. within within The a 3D obstruction a geomtry 3D geomtry modelof seats (seeinmodel the Figure train (see5). isFigure Passengers’ implicitly 5). Passengers’ repres motionented takesmotion as place gapstakes on inplace this the on navigationnavigation this navigation mesh. mesh. mesh. The Since obstruction The the obstruction passengers of seats of can seats in theonly traintravelin the is on implicitly train the isnavigation implicitly represented mesh, repres as thisented gaps technique as in gaps the navigation inprevents the navigation the mesh. overhead mesh. Since ofSince the any passengers thesolid passengers object can representation onlycan only travel travel on the navigation mesh, this technique prevents the overhead of any solid object representation onfrom the affecting navigation the mesh, simulation. this technique prevents the overhead of any solid object representation from affectingfrom affecting the simulation. the simulation.

Figure 5. Navigation mesh of the EMS Maglev train in Pathfinder.

It is worth mentioning that the verification and validation of Pathfinder have been conducted before, e.g., by fundamental diagram tests, both in unidirectional and bidirectional flow and pedestrian behaviour tests at a corridor and stairway intersection [32]. The tests also showed that Pathfinder could provide a good representation of people’s actual movement with respect to a real situation, and it has therefore been widely used in people evacuation research in recent years [33,34]. Sustainability 2019, 11, x FOR PEER REVIEW 6 of 18

Figure 5. Navigation mesh of the EMS Maglev train in Pathfinder.

It is worth mentioning that the verification and validation of Pathfinder have been conducted before, e.g., by fundamental diagram tests, both in unidirectional and bidirectional flow and Sustainability 2019, 11, 6049 6 of 17 pedestrian behaviour tests at a corridor and stairway intersection [32]. The tests also showed that Pathfinder could provide a good representation of people’s actual movement with respect to a real 2.2.2.situation, Evacuation and it Timehas therefore Measurement been widely used in people evacuation research in recent years [33,34].

2.2.2.During Evacuation an interval Time Measurement evacuation, passengers have to transfer from the train carriages to the evacuation walkway as quickly as possible and walk for a long distance, until they reach the safety During an interval evacuation, passengers have to transfer from the train carriages to the exits at the stations. The passengers have different safety risks according to their various positions on evacuation walkway as quickly as possible and walk for a long distance, until they reach the safety the evacuation walkway. For example, in the case of a fire breakout in a Maglev train, passengers are exits at the stations. The passengers have different safety risks according to their various positions on at athe higher evacuation risk if walkway. they are closerFor example, to thetrain in the carriages case of a comparedfire breakout to in the a Maglev passengers train, who passengers have already are evacuatedat a higher to therisk riskif they zone are and closer those to the who train are carriages already outside compared of the to the risk passengers zone [35,36 who]. Jumping have already fire and toxicevacuated gases have to the a higher risk zone tendency and those to harm who those are already who are outside closer of to the risk train zone compared [35,36]. to Jumping those who fire are furtherand toxic away gases from have the traina higher [37]. tendency In addition, to harm passengers those who are are often closer more to atthe risk train of stampedecompared whento those they arewho in the are congested further away connection from the part trai betweenn [37]. In theaddition, Maglev passengers train and theare evacuationoften more walkway.at risk of stampede Hence, it is crucialwhen to they shorten are thein the evacuation congested time connection for passengers, part between especially the in Maglev leaving thetrain train and and the the evacuation evacuation walkwaywalkway. risk Hence, zone it during is crucial the to evacuation. shorten the Measuringevacuation time the evacuationfor passengers, time especially of different in leaving stages the in an evacuationtrain and processthe evacuation might bewalkway helpful risk for improvingzone during the the evacuation evacuation. walkway Measurin design.g the evacuation Therefore, time based onof the different evacuation stages scenario in an evacuation survey, a process new measurement might be helpful of evacuation for improving time the is evacuation proposed, walkway which is as follows:design. Therefore, based on the evacuation scenario survey, a new measurement of evacuation time is proposed, which is as follows: t1—evacuation time, until all passengers have alighted from the train carriages. • • t1—evacuation time, until all passengers have alighted from the train carriages. t2—evacuation time, until all passengers have exited the evacuation walkway risk zone. The total • • t2—evacuation time, until all passengers have exited the evacuation walkway risk zone. The total length of the risk zone is the sum of the total length of the train and the fire separation length on length of the risk zone is the sum of the total length of the train and the fire separation length on both sides of the train—16.5 + 89 + 16.5 = 132 m. The fire separation length is 16.5 m, which is both sides of the train—16.5 + 89 + 16.5 = 132 m. The fire separation length is 16.5 m, which is >12 >12m, m, as asrequired required by bythe the Fire Fire Prevention Prevention Standard Standard for Building for Building Design Design of China of China [38]. [38]. • t —total evacuation time start, until all passengers have passed through the safety exits. • 3 t3—total evacuation time start, until all passengers have passed through the safety exits.

TheThe evacuation evacuation space space division division and and definitionsdefinitions of t1,, tt22 andand t3t 3areare illustrated illustrated in inFigure Figure 6. 6.

FigureFigure 6. 6.Concept Concept of of the the new new evacuationevacuation timetime measurement.measurement. (a ()a )Evacuation Evacuation space space division division of ofan an intervalinterval evacuation evacuation and and (b (b)) the the definitions definitions ofof tt11, t2 andand tt33 evacuationevacuation times. times.

2.2.3.2.2.3. Simulation Simulation Setting Setting The parameters for the evacuation environment in the simulation model are listed in Table1. They were formulated according to the information obtained from the evacuation scenario survey, as mentioned in Section 2.1. The first design parameter of evacuation walkway in the model is the entrance width, d1. According to the evacuation scenario survey, the layout of guardrails determines the entrance width, d1. For example, in the Beijing and Changsha Maglev lines, the guard rail spacing is 0.5 m, which is a typical value in engineering practice to date. Even though the train door width is 1.4 m wide, the actual entrance width of the evacuation walkway remained 0.5 m (provided that the Sustainability 2019, 11, x FOR PEER REVIEW 7 of 18

The parameters for the evacuation environment in the simulation model are listed in Table 1. They were formulated according to the information obtained from the evacuation scenario survey, as mentionedSustainability in 2019Section, 11, 6049 2.1. The first design parameter of evacuation walkway in the model is the 7 of 17 entrance width, d1. According to the evacuation scenario survey, the layout of guardrails determines the entrance width, d1. For example, in the Beijing and Changsha Maglev lines, the guard rail spacing train driver aligned the train door with the entrance) because of a restriction imposed by the guard rail is 0.5 m, which is a typical value in engineering practice to date. Even though the train door width is spacing. A lower bound of 0.5 m was therefore selected for the entrance width, d . Meanwhile, 1.4 m 1.4 m wide, the actual entrance width of the evacuation walkway remained 0.5 m (provided1 that the was selected for the upper bound in order to create room for the improvement of the actual width of a train driver aligned the train door with the entrance) because of a restriction imposed by the guard train door, if necessary. rail spacing. A lower bound of 0.5 m was therefore selected for the entrance width, d1. Meanwhile, Another design parameter is the evacuation walkway width, d . Considering the design 1.4 m was selected for the upper bound in order to create room for the improvement2 of the actual specifications and the actual design conditions, the chosen range for d in this paper was from 0.7 m to width of a train door, if necessary. 2 1.5 m, with 1.1 m as a median value [22]. Considering a 0.1 m interval for both variables, 90 (9 10) Another design parameter is the evacuation walkway width, d2. Considering the design × combinations of d and d were simulated. specifications and the actual1 design2 conditions, the chosen range for d2 in this paper was from 0.7 m to 1.5 m, with 1.1 m as a median value [22]. Considering a 0.1 m interval for both variables, 90 (9 × Table 1. Parameters in the simulation model. 10) combinations of d1 and d2 were simulated. Item Content Table 1. Parameters in the simulation model. Line type Elevated EMS Maglev transit MarshallingItem SixContent carriages LengthLine of each type carriageElevated EMS 16.5 mMaglev transit Door numberMarshalling of each carriageSix Twocarriages Length of each carriage 16.5 m Train door width 1.4 m Door number of each carriage Two Location of evacuation walkway Parallel with the accident train Train door width 1.4 m Location of the evacuation walkway Location of evacuation walkway ParallelAligned with with the carriage accident doors train entrances Location of the evacuation walkway entrances Aligned with carriage doors Location of the safety exits OneOne level level below below the track the track at the stationat the Location of the safety exits Number of safety exitsstation Two NumberInterval of section safety length exits 1500Two m IntervalStation platformsection length length 1001500 m m StationDistance platform to safety length exits 850100 m m Distance to safety exits 850 m Number of staircases in each station One Number of staircases in each station One Staircase width 2.4 m Staircase width 2.4 m

ConnectionConnection between between the evacuation the evacuation walkway walkway and the andtrain the train

0.5 0.6 0.7 0.8 0.9 EvacuationEvacuation walkway walkway entrance entrance width, width, d1 d(m),1 (m), vector 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 vector 1.0 1.1 1.2 1.3 1.4 0.7 0.8 0.9 1.0 1.1 Evacuation walkway width, d2 (m), vector 0.7 0.8 0.9 1.0 1.1 Evacuation walkway width, d2 (m), vector 1.2 1.3 1.4 1.5 — 1.2 1.3 1.4 1.5 — In addition to the parameters in Table 1, the passenger profile, such as the number of passengers, age group,In shoulder addition width to the and parameters moving speed, in Table was1, co thensidered passenger in the profile, model such (see asTable the 2). number The number of passengers, of passengersage group, was shoulder set to 1032, width based and movingon the curre speed,nt wasmaximum considered passenger in the capa modelcity (see for Tableboth2 Beijing). The number Maglevof passengerstransit and wasMaglev set totransit 1032, in based China. on Th thee currentpassengers’ maximum age group passenger consideration capacity from for boththe Beijing questionnairesMaglev transit is summarized and Maglev in Table transit 2. The in China. age group The was passengers’ then utilized age to group estimate consideration the moving from the speedquestionnaires and corresponding is summarized shoulder widt in Tableh, as2. suggested The age group in the was literature then utilized [39]. In to the estimate literature the [40], moving it speed is statedand that corresponding the moving shoulder speed of width,a passenger as suggested in a crowdy in the literaturesituation [can39]. be In different the literature at different [40], it is stated locations,that thee.g., moving the passenger speed ofmovi a passengerng speed on in aa crowdystaircase situation and a slope can is be around different 0.7–0.8 at di fftimeserent slower locations, e.g., in comparison to that on a flat and straight section. Thus, a modification factor of 0.75 was considered the passenger moving speed on a staircase and a slope is around 0.7–0.8 times slower in comparison to for the staircases and connections between the train and the evacuation walkway in the simulation. that on a flat and straight section. Thus, a modification factor of 0.75 was considered for the staircases and connections between the train and the evacuation walkway in the simulation. The pre-action time, the time difference between a person noticing the accident and acting, was also considered in the evacuation time simulation. According to the Society of Fire Protection Engineers (SFPE) handbook of Fire Protecion Engineering, it is suggested that the estimated pre-action time until the evacuation Sustainability 2019, 11, 6049 8 of 17

start shouldSustainabilitySustainability be <120 20192019 s,, 1111 for,, xx FORFOR passengers PEERPEER REVIEWREVIEW who are awake but unfamiliar with the evacuation environment,88 ofof 1818 the alarm system and the evacuation procedure [41]. In this study, to account for realistic passenger The pre-action time, the time difference between a person noticing the accident and acting, was also pre-action time behavior, it was assumed that the pre-action time followed a uniform distribution from consideredconsidered inin thethe evacuationevacuation timetime simulation.simulation. AccordingAccording toto thethe SocietySociety ofof FireFire ProtectionProtection EngineersEngineers 30 s to 120(SFPE)(SFPE) s. handbookhandbook ofof FireFire ProtecionProtecion Engineering,Engineering, itit isis suggestedsuggested thatthat thethe estimatedestimated pre-actionpre-action timetime until the evacuation start should be <120 s for passengers who are awake but unfamiliar with the Table 2. evacuationevacuation environment,environment, thethe alarmalarm systemsystemPassenger andand thethe evacuationevacuation profile. procedureprocedure [41][41].. InIn thisthis study,study, toto accountaccount forfor realisticrealistic passengerpassenger pre-actionpre-action timetime behavior,behavior, itit waswas assumedassumed thatthat thethe pre-actionpre-action timetime Group Age Percentage Shoulder Width Moving Speed followedfollowed aa uniformuniform distributiondistribution fromfrom 3030 ss toto 120120 s.s. Children 15 9% 0.4 m 0.78 m/s ≤ Young person 16–35Table 65% 2. Passenger profile. 0.46 m 1.22 m/s Middle-aged person 36–55 23% 0.46 m 1.17 m/s Group Age Percentage Shoulder Width Moving Speed Elderly people 56 3% 0.46 m 0.75 m/s Children ≥≤≤1515 9% 9% 0.4 0.4 mm 0.78 0.78 m/sm/s YoungTotal person number of 16–35passengers 65% 0.46 m 1032 1.22 m/s YoungMoving person speed modification 16–35 factor 65% 0.46 m 0.75 1.22 m/s Middle-aged person 36–55 23% 0.46 m 1.17 m/s Middle-aged personPre-action 36–55 time 23% Uniform 0.46 distribution m on [30 1.17 s,120 m/s s] Elderly people ≥56 3% 0.46 m 0.75 m/s ElderlyThe people moving mode in≥56 Pathfinder 3% 0.46 m Steering 0.75 m/s Total number of passengers 1032 Moving speed modification factor 0.75 3. Results Pre-action time Uniform distributionribution onon [30[30 s,120s,120 s]s] The moving mode in Pathfinder Steering 3.1. Number of Iterations for the Simulation 3.3. ResultsResults Because of the random effect in the passenger profile in the simulation model, the average evacuation3.1.3.1. time NumberNumber was ofof IterationsIterations used to forfor represent thethe SimulationSimulation a evacuation condition that is closer to reality. Therefore, the proper iterationBecause of times the random for each effect combination in the passenge of therr profileprofile entrance inin thethe width, simulationsimulationd1, andmodel,model, walkway thethe averageaverage width, d2, should beevacuationevacuation considered. timetime waswas To do usedused that, toto representrepresent three combinations aa evacuationevacuation coco ofnditiond1 and thatd is2 closerin bidirectional to reality. Therefore, evacuation the were proper iteration times for each combination of the entrance width, d1, and walkway width, d2, should selected, e.g.,proper the iteration minimum, times for intermediate each combination and maximumof the entrance combinations width, d1, and of walkwayd1 and width,d2. Each d2, should combination be considered. To do that, three combinations of d1 and d2 in bidirectional evacuation were selected, was iteratedbe considered. 50 times. TheTo do randomization that, three combinations function of in d1 Pathfinderand d2 in bidirectional randomized evacuation every were iteration. selected, The mean 1 2 e.g.,e.g., thethe minimum,minimum, intermediateintermediate andand maximummaximum combinationscombinations ofof dd1 andand dd2.. EachEach combinationcombination waswas value of the total evacuation time t of a different number of iterations, K, were compared, and the iteratediterated 5050 times.times. TheThe randomizationrandomization3 functionfunction inin Pathfinder randomized every iteration. The mean 3 differencevalue in the of the percentage total evacuation of the time mean tt3 of evacuation a different number time, t 3of, isiterations, listed in K,,Table were 3compared,. It was observedand the that 3 changingdifference the number in the of percentage iterations of the from mean 5 to evacuation 50 did not time, result tt3,, isis listedlisted in a remarkableinin TableTable 3.3. ItIt waswas diff observedobservederence. thatthat Therefore, changing the number of iterations from 5 to 50 did not result in a remarkable difference. Therefore, 5 5 iterationschanging was selected the number for of each iterations simulation from 5 to combination, 50 did not result and in a the remarkable averages difference. of t1, t2 Therefore,and t3 were 5 used iterations was selected for each simulation combination, and the averages of t1, t2 and t3 were used for for analysis.iterations was selected for each simulation combination, and the averages of t1, t2 and t3 were used for analysis.analysis.

Table 3. t of a3 different iteration number for three combinations. Table3 3. tt3 ofof aa differentdifferent iterationiteration numbernumber forfor threethree combinations.combinations.

3 TheThe Mean ValueValue of of the the Total Total Evacuation AverageAverage Value Value of t (s)of t for3 (s) a Diforff aerent The Mean Value of the Total d1 and1 d2 2 Average Value3 of t (s) for a Case No.Case d1 andand d2 Evacuation Time withwith a a Di Differentfferent K K Combination DifferentNumber Number of Iterations of Iterations Evacuation Time with a Different K No. Combination 𝑻𝑲𝟓𝟎𝑻𝑲𝟓 No. Combination 𝑻𝑲𝟓𝟎𝑻𝑲𝟓T(K=50) T(K=5) K T(K=5) (s) T(K=50) (s) −(%) (%) K T(K=5)T (s)(K =5) (s)T(K=50) (s) T( K=50) (s)𝑻𝑲𝟓𝟎 T (%)) 𝑻𝑲𝟓𝟎 (K=50

d1 = 0.5 m 1 d1 =d0.51 = 0.5 m m 1929.4 1925.3 −0.21% 1 1 2 1929.41929.4 1925.3 1925.3 −0.21% 0.21% d2 =dd0.72 == 0.7 m0.7 mm −

1 d =dd0.91 == 0.9 m0.9 mm 2 2 1 1669.61669.6 1671.5 1671.5 0.11% 0.11% 2 2 1669.6 1671.5 0.11% d2 =dd1.12 == 1.1 m1.1 mm Sustainability 2019, 11, x FOR PEER REVIEW 9 of 18

d =d1.41 = m1.4 m 3 3 1 1522.31522.3 1524.8 1524.8 0.16% 0.16% d2 =d1.52 = m1.5 m

3.2. Bidirectional vs. Unidirectional

The maximum and minimum values of t1, t2 and t3 of 90 combinations for bidirectional and unidirectional evacuation are illustrated in Figure 7. It can be seen that unidirectional evacuation took longer than bidirectional evacuation. This is to be expected, because in unidirectional evacuation, the passengers had only one possible exit. Additionally, altering the combinations of the evacuation walkway design parameters, d1 and d2 resulted in a broader difference between the maximum value and minimum value of t1, t2 and t3 in unidirectional evacuation, compared with bidirectional evacuation.

Figure 7. The maximum and minimum evacuation time of t1, t2 and t3 in two evacuation scenarios of

the 90 combinations of d1 and d2.

Another noticeable phenomenon, which can be seen in Figure 7, is that the difference between the maximum value and the minimum value of t1 and t2 was about 30 s in unidirectional evacuation, while it was about 40 s in bidirectional evacuation. This means that the passengers needed to spend more time in the evacuation walkway risk zone in bidirectional evacuation. The reason is that, in unidirectional evacuation, the passengers did not need to choose exits in the evacuation walkway risk zone. However, in bidirectional evacuation, the passengers were allowed to select any exit during evacuation. Therefore, they could continuously search for the shortest way to evacuate, and this searching behaviour increased the time of evacuation. The evacuation time for carriages, i.e., for all the passengers to get away from each carriage was also a noticeable point. The evacuation time for the passengers to alight from each carriage in all of the simulations (5 × 90 = 450), in both unidirectional and bidirectional evacuations, is shown in the scatter diagram in Figure 8.

Sustainability 2019, 11, x FOR PEER REVIEW 9 of 18

d1 = 1.4 m 3 1522.3 1524.8 0.16% d2 = 1.5 m

Sustainability 2019, 11, 6049 9 of 17

3.2. Bidirectional vs. Unidirectional 3.2. Bidirectional vs. Unidirectional The maximum and minimum values of t1, t2 and t3 of 90 combinations for bidirectional and unidirectionalThe maximum evacuation and minimumare illustra valuested in Figure of t1, t7.2 Itand cant 3beof seen 90 combinationsthat unidirectional for bidirectional evacuation took and longerunidirectional than bidirectional evacuation evacuation. are illustrated This in is Figure to be 7expected,. It can be because seen that in unidirectional unidirectional evacuation evacuation, took the passengerslonger than had bidirectional only one evacuation.possible exit. This Additionally, is to be expected, altering because the combinations in unidirectional of the evacuation,evacuation walkwaythe passengers design had parameters, only one possibled1 and d2exit. resulted Additionally, in a broader altering difference the combinations between the ofmaximum the evacuation value andwalkway minimum design value parameters, of t1, dt12 andandd 2t3resulted in unidirectional in a broader evacuation difference between, compared the maximum with bidirectional value and evacuation.minimum value of t1, t2 and t3 in unidirectional evacuation, compared with bidirectional evacuation.

Figure 7. The maximum and minimum evacuation time of t1, t2 and t3 in two evacuation scenarios of Figure 7. The maximum and minimum evacuation time of t1, t2 and t3 in two evacuation scenarios of the 90 combinations of d1 and d2. the 90 combinations of d1 and d2. Another noticeable phenomenon, which can be seen in Figure7, is that the di fference between the Another noticeable phenomenon, which can be seen in Figure 7, is that the difference between maximum value and the minimum value of t1 and t2 was about 30 s in unidirectional evacuation, while theit was maximum about 40 value s in bidirectional and the minimum evacuation. value This of t means1 and t that2 was the about passengers 30 s in unidirectional needed to spend evacuation, more time whilein the it evacuation was about walkway 40 s in bidire risk zonectional in evacuation. bidirectional This evacuation. means that The the reason passengers is that, needed in unidirectional to spend moreevacuation, time in the the passengers evacuation did walkway not need torisk choose zone exitsin bidirectional in the evacuation evacuation. walkway The risk reason zone. is However, that, in unidirectionalin bidirectional evacuation, evacuation, the passengerspassengers were did allowednot need to to select choose any exits exit during in the evacuation.evacuation Therefore,walkway riskthey zone. could However, continuously in bidirectional search for the evacuation, shortest way the to passengers evacuate, andwere this allowed searching to select behaviour any exit increased during evacuation.the time of evacuation.Therefore, they could continuously search for the shortest way to evacuate, and this searchingThe evacuation behaviour timeincreased for carriages, the time i.e.,of evacuation. for all the passengers to get away from each carriage was also aThe noticeable evacuation point. time The for evacuation carriages, timei.e., for for all the th passengerse passengers to alightto get fromaway each from carriage each carriage in all of was the alsosimulations a noticeable (5 90point.= 450), The in evacuation both unidirectional time for the and passengers bidirectional to evacuations,alight from each is shown carriage in the in scatter all of × thediagram simulationsSustainability in Figure 2019 (5 ,8 ×11. ,90 x FOR = 450), PEER inREVIEW both unidirectional and bidirectional evacuations, is shown 10 of 18in the scatter diagram in Figure 8.

Figure 8. The evacuation time for each carriage for 90 combinations of d1 and d2.(a) Unidirectional Figure 8. The evacuation time for each carriage for 90 combinations of d1 and d2. (a) Unidirectional evacuationevacuation and and (b) bidirectional(b) bidirectional evacuation. evacuation.

In Figure 8a, in unidirectional evacuation, the evacuation time showed a decreasing trend from carriage 1 to carriage 6. Carriage 1, which was the nearest to the left station exit, took the longest to evacuate, while carriage 6, which was the farthest away, took the shortest. In Figure 8b, in bidirectional evacuation, the time was dependent on the location of the carriage, i.e., evacuation in carriage 1 and carriage 6 (nearest to the station exits) took longer, compared to carriage 3 and 4, which lay in the middle of the train. Generally, the closer the train carriage to the safety exit, the longer it took to completely evacuate the passengers. This phenomenon was due to the passenger setting in the simulation, in which the passengers prefered to move and gather closer to the safety exits during the evacuation.

3.3. Evacuation Time vs. Walkway Width and Entrance Width

3.3.1. In Unidirectional Evacuation

Figure 9 illustrates the evacuation times, t1, t2 and t3, with different evacuation entrance width, d1, and walkway width, d2, combinations in unidirectional evacuation.

Sustainability 2019, 11, x FOR PEER REVIEW 10 of 18

Figure 8. The evacuation time for each carriage for 90 combinations of d1 and d2. (a) Unidirectional Sustainability 2019, 11, 6049 10 of 17 evacuation and (b) bidirectional evacuation.

InIn Figure Figure 88a,a, inin unidirectionalunidirectional evacuation,evacuation, thethe evacuation time showed a decreasing trend from carriagecarriage 1 1 to to carriage carriage 6. 6. Carriage Carriage 1, 1, which which was was the the nearest nearest to to the the left left station station exit, exit, took took the the longest longest to to evacuate,evacuate, whilewhile carriagecarriage 6, which6, which was was the farthestthe farthest away, tookaway, the took shortest. the Inshortest. Figure 8Inb, inFigure bidirectional 8b, in bidirectionalevacuation, theevacuation, time was the dependent time was ondependent the location on th ofe thelocation carriage, of the i.e., carriage, evacuation i.e., evacuation in carriage in 1 carriageand carriage 1 and 6 carriage (nearest 6 to(nearest the station to the exits) station took exit longer,s) took comparedlonger, compared to carriage to carriage 3 and 4,3 and which 4, which lay in laythe in middle the middle of the of train. the train. Generally, Generally, the closer the closer the train the train carriage carriage to the to safety the safety exit, theexit, longer the longer it took it took to completely evacuate the passengers. This phenomenon was due to the passenger setting in to completely evacuate the passengers. This phenomenon was due to the passenger setting in the the simulation, in which the passengers prefered to move and gather closer to the safety exits during simulation, in which the passengers prefered to move and gather closer to the safety exits during the evacuation. the evacuation.

3.3.3.3. Evacuation Evacuation Time Time vs. vs. Walkway Walkway Width Width and and Entrance Entrance Width Width

3.3.1.3.3.1. In In Unidirectional Unidirectional Evacuation Evacuation

FigureFigure 99 illustrates illustrates the the evacuation evacuation times, times,t 1t,1t, 2t2and andt 3t,3, with with di differentfferent evacuation evacuation entrance entrance width, width,d1 , dand1, and walkway walkway width, width,d2 ,d combinations2, combinations in in unidirectional unidirectional evacuation. evacuation.

Sustainability 2019, 11, x FOR PEER REVIEW 11 of 18

Figure 9. The evacuation time, tt11, t2 andand tt33, ,for for unidirectional unidirectional evacuation. evacuation. ( (aa)) tt11—evacuation—evacuation time, time, until all passengers had alighted from the train carriages; (b)) tt22—evacuation time, until all passengers had

exited the evacuation walkway risk zone; (c) t3—total—total evacuation evacuation time start, until all passenger passed through the safety exits. The The interpolation interpolation function function,, “interp1”, “interp1”, with with the the meth method,od, “cubic”, “cubic”, in in MATLAB, MATLAB, was used to create the visual eeffectffect of the data changingchanging trend. The black spot represents the average value of the evacuation time for each combination.

From Figure9, it can be seen that the increase in the entrance width, d1, resulted in a fluctuation From Figure 9, it can be seen that the increase in the entrance width, d1, resulted in a fluctuation in the evacuation time of different degrees, when the walkway width was set at different values. When in the evacuation time of different degrees, when the walkway width was set at different values. the evacuation walkway width, d2, was set to 0.7 m and 0.8 m, t1, t2 and t3 increased first and then When the evacuation walkway width, d2, was set to 0.7 m and 0.8 m, t1, t2 and t3 increased first and then decreased with the increase of the entrance width, d1. When d2 was within the range of 0.9–1.5 m, t1, t2 and t3 did not change much, when the entrance width, d1, increased. In addition, increasing the evacuation walkway width, d2, significantly reduced the evacuation time, t1, t2 and t3. The most significant downward trend occurred when the evacuation walkway width was increased from 0.9 m to 1.3 m, whereas only a gentle descent was observed at 0.7–0.8 m and 1.4– 1.5 m, respectively. This is because of the availability of the space on the evacuation walkway. When the evacuation walkway width, d2, was within the range of 0.7–0.8 m, only one row could be formed. However, when d2 was within the range of 0.9–1.3 m, the passengers were able to walk in two rows, which effectively reduced the evacuation time. In the case of d2 being within the range of 1.4–1.5 m, the passengers exhibited a competitive behaviour, i.e., trying to get ahead of each other. Thus, no significant reducing effect on evacuation time was observed when d2 was with the range of 1.4–1.5 m.

3.3.2. In Bidirectional Evacuation

Figure 10 illustrates the evacuation time, t1, t2 and t3, with different evacuation entrance width, d1, and walkway width, d2, combinations in bidirectional evacuation.

Sustainability 2019, 11, x FOR PEER REVIEW 11 of 18

Figure 9. The evacuation time, t1, t2 and t3, for unidirectional evacuation. (a) t1—evacuation time, until all passengers had alighted from the train carriages; (b) t2—evacuation time, until all passengers had

exited the evacuation walkway risk zone; (c) t3—total evacuation time start, until all passenger passed through the safety exits. The interpolation function, “interp1”, with the method, “cubic”, in MATLAB, was used to create the visual effect of the data changing trend. The black spot represents the average value of the evacuation time for each combination.

Sustainability 2019, 11, 6049 11 of 17 From Figure 9, it can be seen that the increase in the entrance width, d1, resulted in a fluctuation in the evacuation time of different degrees, when the walkway width was set at different values. When the evacuation walkway width, d2, was set to 0.7 m and 0.8 m, t1, t2 and t3 increased first and decreased with the increase of the entrance width, d1. When d2 was within the range of 0.9–1.5 m, t1, t2 then decreased with the increase of the entrance width, d1. When d2 was within the range of 0.9–1.5 and t3 did not change much, when the entrance width, d1, increased. m, t1, t2 and t3 did not change much, when the entrance width, d1, increased. In addition, increasing the evacuation walkway width, d2, significantly reduced the evacuation In addition, increasing the evacuation walkway width, d2, significantly reduced the evacuation time, t1, t2 and t3. The most significant downward trend occurred when the evacuation walkway widthtime, t1 was, t2 and increased t3. The most from significant 0.9 m to 1.3 downward m, whereas trend only occurred a gentle when descent the evacuation was observed walkway at 0.7–0.8 width m andwas 1.4–1.5increased m, respectively.from 0.9 m to This 1.3 m, is becausewhereas of only the a availability gentle descent of the was space observed on the at evacuation 0.7–0.8 m walkway. and 1.4– 1.5 m, respectively. This is because of the availability of the space on the evacuation walkway. When When the evacuation walkway width, d2, was within the range of 0.7–0.8 m, only one row could be the evacuation walkway width, d2, was within the range of 0.7–0.8 m, only one row could be formed. formed. However, when d2 was within the range of 0.9–1.3 m, the passengers were able to walk in However, when d2 was within the range of 0.9–1.3 m, the passengers were able to walk in two rows, two rows, which effectively reduced the evacuation time. In the case of d2 being within the range of 1.4–1.5which effectively m, the passengers reduced exhibitedthe evacuation a competitive time. In the behaviour, case of d i.e.,2 being trying within to get the ahead rangeof of each1.4–1.5 other. m, the passengers exhibited a competitive behaviour, i.e., trying to get ahead of each other. Thus, no Thus, no significant reducing effect on evacuation time was observed when d2 was with the range of 1.4–1.5significant m. reducing effect on evacuation time was observed when d2 was with the range of 1.4–1.5 m.

3.3.2. In Bidirectional Evacuation

Figure 1010 illustratesillustrates the the evacuation evacuation time, time,t1 ,t1t,2 tand2 andt3 t,3 with, with di differentfferent evacuation evacuation entrance entrance width, width,d1, andd1, and walkway walkway width, width,d2, d combinations2, combinations in bidirectionalin bidirectional evacuation. evacuation.

Sustainability 2019, 11, x FOR PEER REVIEW 12 of 18

FigureFigure 10.10. The evacuationevacuation time,time,t t11,, tt22 and t3,, for for bidirectional bidirectional evacuation. (a) t1—evacuation time, untiluntil allall passengerpassenger hadhad alightedalighted fromfrom thethe traintrain carriages;carriages; ((bb))t 2t2—evacuation—evacuation time,time, untiluntil allall passengerpassenger hadhad

exitedexited thethe evacuationevacuation walkwaywalkway riskrisk zone;zone; ((cc))t t33—total evacuation time, untiluntil allall passengerpassenger hadhad passedpassed throughthrough thethe safety safety exits. exits. The The interpolation interpolation function, function, “interp1”, “interp1”, with with method, method, “cubic”, “cubic”, in MATLAB, in MATLAB, was usedwas used to create to create the visual the visual effect effect of the of data the changing data changing trend. trend. The black The spotblack represents spot represents the average the average value ofvalue the evacuationof the evacuation time for time each for individual each individual combination. combination.

From Figure 10, it can be seen that, in general, the evacuation time, t1, t2 and t3, fluctuated more, From Figure 10, it can be seen that, in general, the evacuation time, t1, t2 and t3, fluctuated more, comparedcompared toto thosethose inin unidirectionalunidirectional evacuation,evacuation, asas shownshown inin FigureFigure9 9.. The The fluctuations fluctuations caused caused by by the the increase of d1 is more noticeable, when the evacuation walkway width, d2, is equal to 0.7 m, 0.8 m, 1.4 m or 1.5 m, and the opposite is the case when d2 is within the range of 0.9–1.3 m. One possible reason for this might be that when d2 is relatively narrow, the increase in d1 allows more passengers to walk into the risk zone of the evacuation walkway, but the available space in this area is limited. This inconsistency caused a chaotic situation in the simulation, where the passengers gathered around the train door but were not able to evacuate effectively. On the other hand, when d2 is relatively wide, the increase in d1 resulted in more passengers trying to get away from the train and consequently led to a competitive behaviour in the risk zone of the evacuation walkway. In addition, increasing the evacuation walkway width, d2, also significantly reduced the evacuation time, t1, t2 and t3, which was a similar trend to the unidirectional evacuation.

4. Discussion

4.1. The Combined Effect of the Entrance Width and Walkway Width on the Evacuation Time

According to the simulation results of t1, t2 and t3 shown in Figures 9 and 10, it can be observed that there is a combined effect of the entrance width, d1, and the walkway width, d2, on the evacuation time, both in unidirectional and bidirectional evacuation. To quantify the combined effect, the average evacuation time of t1, t2 and t3 with the same d2 in both evacuation scenarios was calculated (see Figure 11). It can be observed that when d2 is increased from 0.7 m to 1.5 m, the average evacuation time of t1, t2 and t3 was reduced by 55.9%, 54.5% and 26.2%, respectively, in unidirectional evacuation, and 36.3%, 35.2% and 17.0%, respectively, in bidirectional evacuation. An increase in the walkway width, d2, resulted in a more significant reduction of the evacuation time in unidirectional evacuation than in bidirectional evacuation.

Sustainability 2019, 11, 6049 12 of 17

increase of d1 is more noticeable, when the evacuation walkway width, d2, is equal to 0.7 m, 0.8 m, 1.4 m or 1.5 m, and the opposite is the case when d2 is within the range of 0.9–1.3 m. One possible reason for this might be that when d2 is relatively narrow, the increase in d1 allows more passengers to walk into the risk zone of the evacuation walkway, but the available space in this area is limited. This inconsistency caused a chaotic situation in the simulation, where the passengers gathered around the train door but were not able to evacuate effectively. On the other hand, when d2 is relatively wide, the increase in d1 resulted in more passengers trying to get away from the train and consequently led to a competitive behaviour in the risk zone of the evacuation walkway. In addition, increasing the evacuation walkway width, d2, also significantly reduced the evacuation time, t1, t2 and t3, which was a similar trend to the unidirectional evacuation.

4. Discussion

4.1. The Combined Effect of the Entrance Width and Walkway Width on the Evacuation Time

According to the simulation results of t1, t2 and t3 shown in Figures9 and 10, it can be observed that there is a combined effect of the entrance width, d1, and the walkway width, d2, on the evacuation time, both in unidirectional and bidirectional evacuation. To quantify the combined effect, the average evacuation time of t1, t2 and t3 with the same d2 in both evacuation scenarios was calculated (see Figure 11). It can be observed that when d2 is increased from 0.7 m to 1.5 m, the average evacuation time of t1, t2 and t3 was reduced by 55.9%, 54.5% and 26.2%, respectively, in unidirectional evacuation, and 36.3%, 35.2% and 17.0%, respectively, in bidirectional evacuation. An increase in the walkway width, d2, resulted in a more significant reduction of the Sustainabilityevacuation 2019 time, 11, in x FOR unidirectional PEER REVIEW evacuation than in bidirectional evacuation. 13 of 18

Figure 11. Average evacuation time of t , t and t for the combinations with the same d in (a) Figure 11. Average evacuation time of t11, t2 2and t3 3for the combinations with the same d2 2in (a) unidirectionalunidirectional evacuation evacuation and and ( (bb)) bidirectional bidirectional evacuation. evacuation.

As shown in Figures9 and 10, it also can be observed that the evacuation time fluctuations were As shown in Figures 9 and 10, it also can be observed that the evacuation time fluctuations were obvious in both the unidirectional and bidirectional evacuation scenarios, when the walkway width obvious in both the unidirectional and bidirectional evacuation scenarios, when the walkway width was 0.7 and 0.8 m. However, in bidirectional evacuation, a more noticeable fluctuation can be observed was 0.7 and 0.8 m. However, in bidirectional evacuation, a more noticeable fluctuation can be when the walkway width was with in the range of 1.4–1.5 m. In general, the increase of d1 resuted observed when the walkway width was with in the range of 1.4–1.5 m. In general, the increase of d1 in the evacuation time varying in different degrees when d2 was set to different values, as well as in resuted in the evacuation time varying in different degrees when d2 was set to different values, as different evacuation scenarios. well as in different evacuation scenarios. Within the range of the two parameters discussed in this article, the combined effect of the entrance Within the range of the two parameters discussed in this article, the combined effect of the width and walkway width on evacuation time may be that increasing the walkway width significantly entrance width and walkway width on evacuation time may be that increasing the walkway width reduces the evacuation time, and the reduction is more noteworthy for unidirectional evacuation than significantly reduces the evacuation time, and the reduction is more noteworthy for unidirectional for bidirectional evacuation. Besides, the increase in the entrance width, d1, resulted in a more obvious evacuation than for bidirectional evacuation. Besides, the increase in the entrance width, d1, resulted evacuation time fluctuation in bidirectional evacuation compared to that in unidirectional evacuation. in a more obvious evacuation time fluctuation in bidirectional evacuation compared to that in unidirectional evacuation.

4.2. Simulation Results Compared to Empirical Formula Results The simulated evacuation time were also compared to the conventional empirical formula of the Code of Safety Evacuation for the Metro of China [42] (see Table 4). In the empirical formula, the total evacuation time, Ttotal, is defined as:

Ttotal = Tpre-a + Ttr + Tpass where Tpre-a—passengers’ pre-action time. Ttr—passengers evacuated from the train. Tpass—passengers evacuated on the passageway. Note that in the conventional empirical formula, the entrance width of the evacuation walkway is not a consideration. Therefore, Ttr is only affected by the walkway width, d2. Besides, Tpass is related to the passenger volume and walking distance on the passageway from the point of evacuation to safety (the station). To make the simulation results comparable to the empirical ones, the average evacuation time, t3, of the simulations with the same walkway width, d2, was used.

Sustainability 2019, 11, 6049 13 of 17

4.2. Simulation Results Compared to Empirical Formula Results The simulated evacuation time were also compared to the conventional empirical formula of the Code of Safety Evacuation for the Metro of China [42] (see Table4). In the empirical formula, the total evacuation time, Ttotal, is defined as:

Ttotal = Tpre-a + Ttr + Tpass

where Tpre-a—passengers’ pre-action time. Ttr—passengers evacuated from the train. Tpass—passengers evacuated on the passageway. Note that in the conventional empirical formula, the entrance width of the evacuation walkway is not a consideration. Therefore, Ttr is only affected by the walkway width, d2. Besides, Tpass is related to the passenger volume and walking distance on the passageway from the point of evacuation to safety (the station). To make the simulation results comparable to the empirical ones, the average evacuation time, t3, of the simulations with the same walkway width, d2, was used.

Table 4. Comparison of the empirical formula and simulation.

Empirical Formula Simulation Evacuation Walkway Tpre-a Ttr Average of Total Evacuation Tpass Tt = Tpre-a + Ttr + Tpass Width d (min) (min) otal Time, t 2 (min) (min) 3 (m) (min) Unidirectional Bidirectional Unidirectional Bidirectional Unidirectional Bidirectional 0.7 2 24.6 19.6 16.2 41.3 37.9 42.4 31.3 0.8 2 21.5 19.6 16.2 38.8 35.4 40.9 29.4 0.9 2 19.1 19.6 16.2 36.9 33.5 39.6 28.9 1.0 2 17.2 19.6 16.2 35.4 32.0 37.9 28.7 1.1 2 15.6 19.6 16.2 34.1 30.7 35.6 28.3 1.2 2 14.3 19.6 16.2 33.1 29.7 33.6 26.9 1.3 2 13.2 19.6 16.2 32.2 28.8 32.1 26.1 1.4 2 12.3 19.6 16.2 31.5 28.1 31.4 26.0 1.5 2 11.5 19.6 16.2 30.8 27.4 31.3 26.0

A decrease in the evacuation time was observed in both the empirical formula and the simulations with the increase of d2. For unidirectional evacuation, the simulation results are similar to the empirical results, and the difference was within the range of 0.1–2.7 min. This is probably because the passengers had only one direction to choose during the evacuation, so they evacuated in a more orderly manner. For bidirectional evacuation, the difference between the simulation results and the empirical results is bigger than that in unidirectional evacuation. This is probably because passenger behavior in bidirectional evacuation, e.g., the process of choosing exits, competitive behavior, etc., was more complex and noticeable. This behavior can be represented in detail with the steering mode in Pathfinder, but it may not be described in detail in the empirical formula. Therefore, the difference between the evacuation time according to the empirical formula and the simulation is slightly larger in the bidirectional evacuation. Besides, in both unidirectional and bidirectional evacuation, although larger walkway widths were set in the simulations, the evacuation time did not show a great improvement, i.e., a walkway width of 1.3–1.5 m did not cause a drastic decrease in the evacuation time. Generally, the empirical formula provides more conservative results. However, it still might be concluded that the empirical formula and the Pathfinder are in better agreement regarding the prediction of the evacuation time in unidirectional evacuation. However, in bidirectional evacuation, or a more complicated evacuation, a numerical simulation with the capability of including passenger behavior might provide more detailed insight. In addition, the new evacuation time measurement method, t1, t2 and t3, proposed in this study could be seen as a novel trial for evacuation time study in relation to elevated rail transit systems. Sustainability 2019, 11, 6049 14 of 17

One advantage of this method is that the evacuation time and the evacuation process on the elevated lines can be analyzed more accurately and flexibly and be understood more intuitively. For example, t1 (evacuation time, until all passengers had exited the train carriage zone) is helpful to find the most crowded and the most evacuation time costly carriages. In this paper, it is revealed that the most congested carriages in interval evacuation are those closest to the safety exits in the station, i.e., both ends of the train. The other advantage of this measurement is that t2 (evacuation time, until all passengers had exited the evacuation walkway risk zone) is a more suitable measurement reference, compared to Ttotal, in studying evacuation walkway design. This is because, when a train is caught in an accident on an elevated railway line, the dangerous area is more likely to be around the train. Thus, shortening the evacuation time, t2, for passengers to leave the evacuation walkway risk zone is more important than the whole evacuation time.

4.3. Suggestions for Evacuation Walkway Parameter Design

As discussed above, combinations of d1 and d2 corresponding to the shortest t2 in both directions were listed in Table5. As shown in the results, for both evacuation directions, t2 decreased dramatically with the increase of d2, until the width of the walkway was 1.3 m, but an increase of d2 greater than 1.3 m did not cause a significant reduction in t2. Therefore, in this study, it might be safe to assume that a d2 of 1.3 m is an ideal width for new elevated transit line evacuation walkways, and the entrance width, d1, is suggested to be 1.4 m, as this combination might potentially help to reduce the evacuation time, t2, from 1295 s to 814 s, reduced by 37.2%, compared to the current design practice (d1 = 0.5 m, and d2 = 1.0 m) in the unlikely event of unidirectional evacuation. For a restricted construction condition, where the evacuation walkway width, d2, is 0.7 m, the evacuation time, t2, may potentially be reduced from 1606 s to 1485 s by changing the entrance width of 0.5 m to 1.4 m. Meanwhile, for the evacuation walkway width, d2 = 1.0 m, which is currently the normal practice, the evacuation time, t2, may potentilly be reduced from 1295 s to 1278 s by changing the entrance width of 0.5 m to 1.4 m. Generally, a broader entrance width that is 1.4 m (the train door width in this study) or as wide as possible is suggested.

Table 5. Combinations of d1 and d2 corresponding to the shortest t2.

Unidirectional Evacuation Bidirectional Evacuation Evacuation Walkway Width, d2 (m) Entrance Width, The Minimum Entrance Width, The Minimum d1 (m) Value of t2 (s) d1 (m) Value of t2 (s) 0.7 1.4 1485 1.0 791 0.8 0.6 1411 0.8 772 0.9 0.5 1386 0.8 720 1.0 1.4 1278 0.8 692 1.1 0.8 1151 1.2 633 1.2 1.2 927 1.0 577 1.3 1.4 814 1 531 1.4 1.4 756 1.2 516 1.5 0.9 717 1.2 511

The limitations of this study are as follows:

Only one simulation setting, namely, evacuation in the middle of an interval section on an elevated • transit line was considered. This setting considered the most likely worst-case scenario during an emergency on an elevated transit line, i.e., with the longest walking distance on both sides to safety exits. In a highly unlikely event, a train might stop close to the station, but only one-way escape is available. This study has also partly covered unidirectional evacuation up to point t2 but will have a different t3. Sustainability 2019, 11, 6049 15 of 17

The total number of evacuation passengers was not set as a variable, but only a maximum number • of 1032 is used. This is assumed to be the worst-case evacuation scenario, as fewer people would result in a less chaotic situation. Six carriage marshalling, with a 1032 passenger capacity, was considered in this study, because it is the current maximum passenger capacity and maximum carriage marshalling for a Maglev train operation. The number of carriages and passengers can be set as variable as longer trains might be anticipated in the future.

5. Conclusions This paper contributed to the improvement of passenger evacuation conditions on elevated rail transit lines. Based on literature research and an evacuation scenario survey, a new evacuation time measurement for interval evacuation on elevated rail transit was first introduced. Then, an interval evacuation model on elevated rail transit lines was built using a multi-agent simulator, Pathfinder, with qualitative and quantitative data that were surveyed and collected earlier. After that, the model was used to study the evacuation time for different combinations of evacuation walkway design parameters, namely, the walkway entrance width and walkway width in both unidirectional and bidirectional evacuation. The following conclusions were drawn: There is a combined effect of the evacuation walkway width and entrance width on the evacuation time in both unidirectional and bidirectional evacuation. Based on the conditions set in the model, it is observed that an increase in the walkway width from 0.7 m to 1.5 m may potentially reduce the evacuation time, t2, by 54.5%, in unidirectional evacuation, and 35.2% in bidirectional evacuation. The increase in the walkway entrance width did not cause a great improvement in reducing the evacuation time, but the incease of the entrance width can result in evacuation time fluctuation. The evacuation time fluctuation is more obvious in bidirectional evacuation, compared to unidirectional evacuation. A reasonable match between the walkway entrance width and walkway width can significantly reduce evacuation time. Therefore, the walkway entrance width and walkway width should be considered together to achieve a shorter evacuation time in evacuation walkway design. Our suggestion for a new evucation walkway design is 1.3 m and 1.4 m for the walkway width and entrance width, respectively. This may potentially reduce the passenger evacuation time from the risk zone by 37.2%, compared to the current normal design practice, where the walkway width is 1.0 m, and the entrance width is 0.5 m. In the case of limited space for evacuation walkway construction, 0.7 m and 1.4 m for the walkway width and entrance width, respectively, should be adopted. For existing evacuation walkways, which have a typical width of 1 m, enlargement of the entrance to 1.4 m is recommended.

Author Contributions: Conceptualization, Z.P. and Q.W.; methodology, Z.P., Q.W. and O.T.; software, Z.P. and A.L.; validation, Z.P.; formal analysis, Z.P., Q.W. and A.L.; investigation, Z.P. and Q.W.; data curation, Z.P. and Q.W.; writing—original draft preparation, Z.P., Q.W., O.T. and A.L.;writing—review and editing, Z.P., O.T. and A.L.; visualization, Z.P. and A.L.; supervision, Z.P., Q.W. and O.T.; project administration, Z.P., Q.W. and O.T.; funding acquisition, Q.W. and O.T. All authors reviewed the results and approved the final version of the manuscript. Funding: This research was funded by the “BEIJING NATURAL SCIENCE FOUNDATION, grant number 8172040”, “A SCHOLARSHIP UNDER THE STATE SCHOLARSHIP FUND OF P.R CHINA” and the “NORWEGIAN UNIVERSITY OF SCIENCE AND TECHNOLOGY, grant number 649105”. Acknowledgments: The authors are grateful to the Beijing Natural Science Foundation (Grantno.8172040), the China Scholarship Council and the Norwegian University of Science and Technology for their financial support. The authors are also grateful to Wenjun Lu and Diego Barbieri for providing good advice for the improvement of this paper. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. Sustainability 2019, 11, 6049 16 of 17

References

1. Xue, H.; Yang, P.; Zhang, H.; Jing, E. Study on the Control Strategy of Urban Rail Transit Passenger Flow under the Condition of Large Passenger Flow. IOP Conf. Ser. Earth Environ. Sci. 2019, 234, 012001. [CrossRef] 2. Burnett, J. Fire Safety Concerns for Rail Rapid Transit Systems. Fire Saf. J. 1984, 8, 3–7. [CrossRef] 3. Kyriakidis, M.; Hirsch, R.; Majumdar, A. Metro Railway Safety: An Analysis of Accident Precursors. Saf. Sci. 2012, 50, 1535–1548. [CrossRef] 4. Gershon, R.R.M.; Qureshi, K.A.; Barrera, M.A.; Erwin, M.J.; Goldsmith, F. Health and Safety Hazards Associated with Subways: A Review. J. Urban Health 2005, 82, 10–20. [CrossRef] 5. Ouyang, M.; Hong, L.; Yu, M.-H.; Fei, Q. STAMP-based Analysis on the Railway Accident and Accident Spreading: Taking the China–Jiaoji Railway Accident for Example. Saf. Sci. 2010, 48, 544–555. [CrossRef] 6. Akabal, F.M.; Masirin, M.I.H.M.; Akasah, Z.A.; Rohani, M.M. Review on Selection and Suitability of Rail Transit Station Design Pertaining to public safety. In Proceedings of the IOP Conference Series: Materials Science and Engineering, Melaka, Malasia, 6–7 May 2017; Volume 226, p. 012033. 7. Vuilleumier, F.; Weatherill, A.; Crausaz, B. Safety aspects of railway and road tunnel: Example of the Lötschberg railway tunnel and Mont-Blanc road tunnel-ScienceDirect. Tunn. Undergr. Space Technol. 2002, 17, 153–158. [CrossRef] 8. Cheng, H.; Yang, X. Emergency Evacuation Capacity of Subway Stations. Procedia Soc. Behav. Sci. 2012, 43, 339–348. [CrossRef] 9. Hunter-Zaworski, K. Analysis of Passenger Incident Data from Five Rail Transit Systems. Safety 2017, 3, 21. [CrossRef] 10. Lombardi, M.; Cuarascio, M.; Rossi, G. The Management of Uncertainty: Model for Human Error Probability in Railway System. Am. J. Appl. Sci. 2014, 11, 381–390. [CrossRef] 11. Seike, M.; Kawabata, N.; Hasegawa, M. Experiments of Evacuation Speed in Smoke-filled Tunnel. Tunn. Undergr. Space Technol. 2016, 53, 61–67. [CrossRef] 12. Benn, J. Railway Bridge Failure during Flooding in the UK and Ireland. Proc. Inst. Civ. Eng. Forensic Eng. 2013, 166, 163–170. [CrossRef] 13. Kucera, P.; Bradáˇcová, I. Modelling the Evacuation of People from a Train on Fire in a Railway Tunnel. In Recent Advances in Engineering. 3rd European Conference of Chemical and Civil Engineering; Angrey, D., Catarina, L.C., Eds.; WSEAS Press: Paris, France, 2012; pp. 196–201. 14. Chen, S.-K.; Li, S.-Y.; Li, X.; Hong, J.; Lai, J.-X.; Cai, P. Modeling Evacuation Time for Passengers from Metro Platforms. J. Transp. Syst. Eng. Inf. Technol. 2008, 8, 101–107. 15. Li, Y.; Liu, S. Passenger Evacuation in Subway Transit Tunnels Installed with Side Evacuation Platforms. Urban Rapid Rail Transit 2007, 20, 13–16. 16. Zhai, W.; Zhao, C. Dynamics of Maglev Vehicle/Guidway Systems(I)-Magnet/Rail Interaction and System Stability. Chin. J. Mech. Eng. 2005, 41, 1–10. [CrossRef] 17. Fitting Out The Railway. Available online: http://www.crossrail.co.uk/construction/railway-systems/ (accessed on 24 September 2019). 18. Habicht, A.T.; Braaksma, J.P. Effective Width of Pedestrian Corridors. J. Transp. Eng. 1984, 110, 80–93. [CrossRef] 19. Lundström, F.V.; Ahlfont, J.; Nilsson, D. The Effect of Raised Walkway Design on Evacuation Behaviour in Rail Tunnels. Fire Saf. Sci. 2014, 11, 1091–1102. [CrossRef] 20. European Commision. Commision Decision of 20 December 2007. Concerning the Technical Specification of Interoperability Relating to ‘Safety in Railway Tunnels’ in the Trans-European Conventional and High-Speed Rail System. 2008. Available online: https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CONSLEG: 2008D0163:20130124:EN:PDF (accessed on 31 October 2019). 21. Standardization Administration of the People’s Republic of China. Code for Design of Metro; Standardization Administration of the People’s Republic of China: Beijing, China, 2013; pp. 26–27. 22. Minishty of Housing and Urban-Rural Developement of the People’s Republic of China. Code for Design of Medium and Low Speed Maglev Transit; Minishty of Housing and Urban-Rural Developement of the People’s Republic of China: Beijing, China, 2017. 23. Helbing, D.; Farkas, I.J.; Molnar, P.; Vicsek, T. Simulation of pedestrian crowds in normal and evacuation situations. Pedestr. Evacuation Dyn. 2002, 21, 21–58. Sustainability 2019, 11, 6049 17 of 17

24. Fridolf, K.; Nilsson, D.; Frantzich, H. The flow rate of people during train evacuation in rail tunnels: Effects of different train exit configurations-ScienceDirect. Saf. Sci. 2014, 62, 515–529. [CrossRef] 25. Cuesta, A.; Abreu, O.; Balboa, A.; Alvear, D. An experimental data-set on merging flows in rail tunnel evacuation. Tunn. Undergr. Space Technol. 2017, 70, 155–165. [CrossRef] 26. Railway Bureau; Ministry of Land, Infrastructure, Transport and Tourism. Technical Regulatory Standards on Japanese Railways; Railway Bureau; Ministry of Land, Infrastructure, Transport and Tourism: Tokyo, Japan, 2012; p. 196. 27. Reynolds, C.W. Steering behaviors for autonomous characters. In Proceedings of the Game Developers Conference, San Jose, CA, USA, 15–19 March 1999; Volume 1999, pp. 763–782. 28. Heni, A.; Jan, M.; Oliver, O. Fast, Neat, and Under Control: Arbitrating Between Steering Behaviors. AI Game Program. Wisdom 2006, 3, 221–232. 29. Thornton, C.; O’Konski, R.; Hardeman, B.; Swenson, D. Pathfinder: An Agent-Based Egress Simulator. In Pedestrian and Evacuation Dynamics; Peacock, R.D., Kuligowski, E.D., Averill, J.D., Eds.; 2011; pp. 889–892. 30. Thornton, C.; O’Konski, R.; Klein, B.; Hardeman, B.; Swenson, D. New Wayfinding Techniques in Pathfinder and Supporting Research. In Pedestrian and Evacuation Dynamics 2012; Weidmann, U., Kirsch, U., Schreckenberg, M., Eds.; Springer International Publishing: Berlin/Heidelberg, Germany, 2014; pp. 1315–1322. 31. Thunderhead Engineering. Technical Reference of Pathfinder. Available online: https://www.thunderheadeng. com/pathfinder/resources/ (accessed on 28 September 2019). 32. Thunderhead Engineering. Pathfinder Vertification and Validation. Available online: https://www. thunderheadeng.com/pathfinder-verification-validation-archive/ (accessed on 28 September 2019). 33. Ronchi, E.; Colonna, P.; Capote, J.; Alvear, D.; Berloco, N.; Cuesta, A. The Evaluation of Different Evacuation Models for Assessing Road Tunnel Safety Analysis. Tunn. Undergr. Space Technol. 2012, 30, 74–84. [CrossRef] 34. Wang, H.; Chen, Q.; Yan, J.; Yuan, Z.; Liang, D. Emergency Guidance Evacuation in Fire Scene Based on Pathfinder. In Proceedings of the 2014 7th International Conference on Intelligent Computation Technology and Automation, Changsha, China, 25–26 October 2014; pp. 226–230. 35. White, N.; Dowling, V.P.; Barnett, J.R. Full-scale Fire Experiment On A Typical Passenger Train. Fire Saf. Sci. 2005, 8, 1157–1168. [CrossRef] 36. Chiam, B.H. Numerical Simulation of a Metro Train Fire. Master’s Thesis, University of Canterbury: Christchurch, Canterbury, New Zealand, 2005. 37. Linteris, G.T.; Rafferty, I.P. Flame size, heat release, and smoke points in materials flammability. Fire Saf. J. 2008, 43, 442–450. [CrossRef] 38. Minishty of Housing and Urban-Rural Developement of the People’s Republic of China. Code for Fire Protection Design of Buildings; Minishty of Housing and Urban-Rural Developement of the People’s Republic of China: Beijing, China, 2018. 39. Wenting, L. Research on Passengers Capacity of Emergency Evacuation in Subway. Master’s Thesis, Tongji University, Shanghai, China, 2008. 40. Research on Emergency Evacuation Behaviour of Passengers in Subway Station Fire. Master’s Thesis, Hunan University of Science and Technology, Hunan, China, 2012. 41. Hurley, M.J.; Gottuk, D.T.; Hall, J.R., Jr.; Harada, K.; Kuligowski, E.D.; Puchovsky, M.; Watts, J.M., Jr.; WIECZOREK, C.J. SFPE Handbook of Fire Protection Engineering; Springer: Berlin/Heidelberg, Germany, 2015; ISBN 978-1-4939-2565-0. 42. Standardization Administration of the People’s Republic of China. Code for Safety Evacuation of Metro; Standardization Administration of the People’s Republic of China: Beijing, China, 2017; pp. 8–10.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).