OriginalScientificArticle RestartofHospitalityandTourism:SystemDynamics andScenario-BasedModelling PetrŠtumpf PragueUniversityofEconomicsandBusiness petr.stumpf@vse.cz JitkaMattyašovská PragueUniversityofEconomicsandBusiness jitka.mattyasovska@seznam.cz AdrianaKrištůfková PragueUniversityofEconomicsandBusiness kria02@vse.cz Atourismdestinationisdefinedasanopen,complex,andadaptivesystem,inwhich numerous relations in the economic, social, and environmental spheres are gener- ated. This paper aims to define a system dynamics model of tourism destinationas a complex system and to identify future behaviour of the system after the restartof tourism in the post-covid-19 era. The main methodological approaches were sys- tem dynamics and simulation modelling. The case of a complex tourism system in the South Bohemia Region, the Czech Republic, in the form of a Stocks and Flows Diagram (sfd) is presentedin this paper, focusing on the business activities at this tourism destination. The simulation results show the future behaviours of the sys- teminvariousscenariosandcomparethedevelopmentofseveraleconomicindica- tors. Three possible future scenarios of a restart of the hospitality and tourism in- dustry are compared with the theoretical situation without covid-19 disease. The proposedsystemdynamicsmodelcontributestothecurrenttheoryoftourismdes- tination management systems and can be used practically by destination managers fordestinationplanningandtoformulatedestinationstrategies. Keywords: systemdynamics,simulationmodelling, tourismdestination, destinationmanagement https://doi.org/10.26493/2335-4194.14.125-136 Introduction A tourism destination system involves a great num- berofstakeholders.Oneofthemostsignificantstake- holders are the tourism enterprises that are regarded as a ‘backbone’ of the tourism destination system. A destinationinwhichtourismenterprisesoperatehasa significantimpactonthecompetitivenessoftheseen- terprises and their performance. However, the oppo- siterelationalsoapplies.Itmeansthatthecompetitive- nessofthedestinationisnoticeablydependentonthe competitiveness of the enterprises in the destination, in terms of each individual company and all compa- niesinaggregate(Ritchie,2003). The ability to compete in the tourism market is, from the perspective of individual entrepreneurs, the subject of their interest; on the other hand, the com- Academica Turistica, Year 14,No. 2,December 2021 |125 Petr Štumpfetal. Restart of Hospitality and Tourism petitiveness of the whole industry and aggregated re- sults of the private sector in the destination are im- portantfor the public administration.Thus, the com- petitiveness of the whole destinationshould be in the spotlight for destination management as represented by destinationmanagementorganisation(dmo). The hospitality and tourism industry has suffered enormously from the covid-19 pandemic and gov- ernment restrictions in all countries. The behaviour ofthewholetourismsysteminthepost-covid-19pe- riodisstillunclear,aswell. Therefore, the main ambition of this paper is to define a system dynamics model of tourism destina- tionasacomplexsystemandtosimulatepossiblesce- nariosoffuturedevelopmentafterthetourismsystem restart.WeusethecaseoftheSouthBohemiaRegion. TheSouthBohemiaRegionrepresentsoneofthemost populartouristregionsintheCzechRepublic,rightaf- ter the capital city of Prague and the South Moravia Region. The aim is to provide a practical tool in the formofacomplexmodel,whichcouldbeusedbydes- tination managers to facilitate their decision making, destination planning, and destination strategies for- mulationin post-covid-19 tourismdevelopment. Therefore, we formulate the following research question:Howwillthehospitalityandtourismindustry developinthepost-covid-19eraintheSouthBohemia Region? Weusesystemdynamicsasthemainmethodolog- ical approach to answerthe postulated researchques- tion. A tourism destination is considered a dynamic complex system because it comprises many different components that interact in a non-linear way (Bag- gio & Sainaghi, 2011; Mai & Smith, 2018) and, there- fore, it needs to be appropriately modelled to achieve efficient destination management (Bieger, 2008; Far- rell&Twining-Ward,2004;Lew&McKercher,2006; Rodriguez-Diaz & Espino-Rodriguez, 2007). System dynamicsisamethodtoenhancelearningincomplex systemswhichoftenusescomputersimulationmodels tohelpuslearnaboutdynamiccomplexityanddesign more effective policies (Sterman, 2000). This method can be understood as a computer-based approach to understandandanalyseasystem’sbehaviourovertime (Sedarati et al., 2019). Therefore, we use system dy- namics to simulate possible scenarios, because future tourism development in the post-covid-19 period is stillunclearandwill requirecomplexsolutions. The proposedsystemdynamicsmodelcontributes to the current theory of tourism destinationmanage- mentsystems.Systemdynamicsin travelandtourism researchisusedbyotherresearchersaswell(Borštnar etal.,2011;JereJakulin,2016;2017;JereLazanski&Kl- jajic, 2006; Mai & Smith, 2018; Patterson et al., 2004; Ropret et al., 2014; Sedarati et al., 2019; Štumpf & Vo- jtko,2016;Tegegneetal.,2018;Vojtko&Volfová,2015). However, the previous studies do not include such a highnumberofvariablesandinterrelationsanddonot cover the complexity of the whole destination system asdoes thepresentedmodel. Only a few authorssim- ulatefuturescenarios(Mai&Smith,2018).Thus,we see the gap in the theory and provide a scientific tool forfuturedirectionsoftourisminthesechaotictimes. TheoreticalBackground The use of the systemic approach in tourism origi- natesfrom the fact thattourism destinationsarecon- sidered complex systems (Baggio & Sainaghi, 2011; Kaspar, 1976; Laesser & Beritelli, 2013; Mai & Smith, 2018;Štumpf&Vojtko,2016).AccordingtotheSankt- Gallen consensus of destination management, desti- nationscanbeunderstoodnotonlyasgeographicen- tities,clustersornetworksofsuppliersbutalsoaspro- ductivesocial systemswithspecificbusinessaimsand non-businessrelatedgoals(Laesser&Beritelli,2013). TourismDestinationasaComplexSystem S ystemstheo ryisusedaso neo ftheessen tiala p- proaches to studying and managing the travel and tourism industry (Kaspar, 1976), especially in a spe- cific environment of tourism destinations. Based on this theory, a tourism destination is defined as an open,complex andadaptivesystem,in whichnumer- ous relations in the economic, social, and environ- mental spheres are generated. A tourism destination is considered a dynamic complex system because it comprises many different components that interact in a non-linear way (Baggio & Sainaghi, 2011; Mai & Smith, 2018). The tourism destination as a complex system needs to be appropriately modelled to achieve 126 | Academica Turistica, Year 14,No. 2,December 2021 PetrŠtumpf etal. Restart of Hospitality and Tourism efficient destination management (Bieger, 2008; Far- rell&Twining-Ward,2004;Lew&McKercher,2006; Rodriguez-Diaz & Espino-Rodriguez,2007). The system also contains many stakeholders with entirelydifferentmanagementobjectivesandinterests (Mai & Smith, 2018; Štumpf & Vojtko, 2016), and it is influenced by various internal factors (such as pol- icy,governmentregulations,andsocio-economiccon- ditions) as well as external factors (such as the eco- nomic situation, safety and security, and technologi- cal or environmental changes). It means that manag- ingatourismdestinationisuncertain,anddestination managers have to make decisions in a complex envi- ronment(Mai &Smith,2018). TourismDestinationandSystemDynamics The first system dynamics models were used for sim- ulations in businesses (Forrester, 1961). However, sys- tem dynamics modelling enables evaluating the eco- nomic impacts and the socio-cultural and environ- mental impacts and their mutual interactions (Jack- son, 2003). In comparison to other methods that are often used for the evaluation of the economic im- pactoftourismondestinations,systemdynamics has one advantage – it can be operated at the same time with ‘soft’ factors from the social and environmental spheres, non-linear relations, delays, and causal loops (reinforcing or balancing), in one complex model (Sterman, 2000). Thus, we can observe stakeholders and general tourism development in destinationsin a broadercontextwiththeemphasisonsustainability. The system dynamics searches for an explanation ofphenomena(variableswithintheboundariesofthe system).Theendogenousapproachcreatessystemdy- namicsthroughtheinteractionofvariablesandagents represented in the model. By specifying the structure of the system and the rules of interaction (decision- making rules in the system), it is possible to reveal behaviour patterns created on the basis of these rules andthisstructure,andtodiscoverhowbehaviourcan be changed following the alternation of the structure andrules(Sterman,2000).Forexample,JereLazanski and Kljajic (2006) or Mai and Smith (2018) have used thisapproachin dynamic modelling of tourismdesti- nations. In contrast, the approach based on the exogenous variables(variablesbeyondthemodelboundaries)ex- plains the dynamics of given variables in the sense of othervariableswhosebehaviourisanticipated.Anen- dogenous explanation of the system dynamics does not mean that the models should never contain any exogenousvariables.However,thenumberofexternal inputs should not be high, and each ‘exogenous input candidate’ must be carefully verified. Careful consid- eration must be given to whether there is significant feedbackfromendogenouselementstotheconsidered exogenousinputinthesystem.Ifso,theboundariesof thesystemmustbeextended,andthisvariablemustbe modelledasendogenous(Sterman,2000). An approach based on exogenous variables has been used in tourism by, for example, Patterson et al. (2004), who deal with a dynamics system of sustain- abletourismontheCaribbeanislandofDominica.At first, the authors identified exogenous variables such as the global economy, politics, and climatic condi- tions.Onlythendidtheyoutlinethreebroadendoge- nous areas of research in which they identified indi- vidualvariables–society(population,migration,etc.), ecosystem (land exploitations, portable capacity, etc.) andeconomics (gdp, income fromtourism,etc.). Severalresearchstudieshavebeenpublishedinthe field of travel and tourism, using system dynamics as the main theoretical approach (Borštnar et al., 2011; Jere Jakulin, 2016; 2017, 2019; Jere Lazanski & Kljajic, 2006; Mai & Smith, 2018; Patterson et al., 2004; Ro- pretetal.,2014;Sedaratietal.,2019;Štumpf&Vojtko, 2016; Tegegne et al., 2018; Vojtko & Volfová, 2015). Moreover, Schianetz et al. (2007), based on Senge’s (1990) theory of Learning Organization, present the concept of Learning Tourism Destination using sys- temdynamicsasatoolforimplementingandreinforc- ingcollectivelearningprocesses.Theresultsshowthat system dynamics methodology can support commu- nication among crucial stakeholdersin tourism desti- nationsandstimulateorganisationallearning. SimulationModellinginTourismResearch Modelling in tourism is used mainly to understand complex systems and connections when, on the basis of the clarification of certain phenomena, it is possi- Academica Turistica, Year 14,No. 2,December 2021 |127 Petr Štumpfetal. Restart of Hospitality and Tourism bletoimitatethebehaviouroftheinvestigatedsystem, simulate it on the specific model, and then influence its behaviour. Simulation models are used in tourism, forexample,topredictsupplyanddemand,determine the impact of tourismon the economy, the local com- munities and the environment, to model movement of tourists in the destination, or as a tool facilitating decision-making in planning and defining develop- ment and marketing strategies (Ahlert, 2008; Ander- gassenetal.,2013;Athanasopoulos&Hyndman,2008; Bonhametal.,2009;Buchta&Dolnicar,2003;Greiner, 2010; Lacitignola et al., 2007; Lawson, 2006; Lew & McKercher,2006; Liu etal.,2012). Nowadays, computer simulations are increasingly usedinsocialsciencesasatoolforunderstandingvar- ious social phenomena.Employing simulation, scien- tistscandeterminecausaleffects,specifykeyparame- terestimates,andclarifytheevolutionoftheprocesses over time. In addition, simulation methods are often very effective in terms of time and costs; sometimes, they are even the only possible means for examining certainphenomena(Garson,2008).Themainareasof simulations used in the social sciences aresystem dy- namics models, network models, spatial models, and agent-basedmodels. Focusing on this research study, simulations gro- undedinsystemdynamicscouldbeusedtobetterun- derstand the structure of the complex tourism desti- nationsystemanditsbehaviour inatime perspective. These simulations can combine many different inter- related factors and play an important role in testing various scenarios. That is why such system dynamics simulationmodelscanbeusedtomakestrategicdeci- sionsandforstrategicplanningintourismdestination developmentingeneral. Methods The main methodological approach was system dy- namics modelling. In line with the previous studies, we built the model based on system dynamics mod- elling,accordingtothesystemdynamicsmethodology (Jackson,2003).Thefirststepconsistsofidentifyinga research problem and variables, which have a crucial influenceonthedefinedproblem.Thevariablescreate theboundariesofthesystem. TheStocksandFlowsDiagramConstruction The presented system dynamics model in the form ofaStocksandFlowsDiagram(sfd)showsthein- teractions among the defined variables and reveals th eco m p lexs truct ur eo fth em ode l .J er eLaza n s ki and Kljajic (2006) defined the relations among the model, the object, and the modelling subject. Based on this approach, the object of the model was defined asthe dynamics of tourismdevelopmentin the South Bohemia Region. The subject of the model is then represented by the researchers (authors) as the ob- servers/descriptors of the model. The sfd represents a mathematical simulation model. Figure 1 shows the sfd structure. The compiled model of the tourism destination system includes 14 stock variables that form the base ofthemodel.Eachstockvariablehasitsowninflow(s) andusually,butnotnecessarily,outflow(s).Stocksrep- resent accumulations within a system and flows in- crease (inflows) or decrease (outflows) stocks. Auxil- iary variables and stocks control the flows. Therefore, a stock can be changed only via its flows, and stocks andauxiliaryvariablescontroltheflows(Mai&Smith, 2018). Constants are used for setting the policies and scenarios simulations. Figure 2 shows a part of the sfd focusing on accommodation capacity where the Accommodationestablishments(ae)capacity variable represents the stock,Investments the inflow,aeclos- ing andDepreciation the outflows,aeoccupancy and aebuildingnecessity the auxiliary variables, andAd- ditionalinvestments theconstant. In this study, we focus primarily on the variables linked to the entrepreneurs’ performance, such as Profit&Loss, accommodation capacity, or number of daysspentbyvisitorsinthedestination.However,the model enables us to set the policies and simulate sce- nariosinasustainablemannerbecauseitincludesalso the variables related to public administration(e.g. tax revenues), residents’ attitudes (residents irritation), and the environment (cultural and natural potential). Themodel structureis describedin Appendix 1. ModelCalibrationandValidation Afterthe sfd structureconstruction,themodelmust becalibratedwithparametervaluestorunthesimula- 128 | Academica Turistica, Year 14,No. 2,December 2021 PetrŠtumpf etal. Restart of Hospitality and Tourism Figure1 Tourism Destination System: Stocks andFlows Diagram tions.Theseparametersinclude(a)theinitialvaluefor stocksatthebeginningofthesimulation,(b)constants thatarestoredasauxiliaryvariables,and(c)graphical functions that represent the influence of one variable onanother.Theremainderofthe sfd isparametrised using equations (Mai & Smith, 2018; Sterman, 2000). The time step of the simulation is one month, and the simulationsrunfor120time-steps(10years). Academica Turistica, Year 14,No. 2,December 2021 |12 9 Petr Štumpfetal. Restart of Hospitality and Tourism Figure2 Stocks andFlows Diagram: Accommodation Capacity A wide set of secondary data about the numbers of destination visitors, length of stay, and accommo- dation capacity was collected to calibrate the simula- tionmodel.Thebaseyearforthesestatisticswas2019. Some variables, such as the price level, indications of quality, satisfaction, or residents’ irritation, were esti- matedbasedonconsultationswithprofessionalsfrom theregion. Calibration of the simulation model, as well as the initialvalues,equationsanddatasourcesareshownin the supplementaryfile generatedby Vensim 6 Profes- sional(https://fm.vse.cz/english/sfd-irritation2). We validated the simulation model to achieve the real-lifebehaviourofthesystem.Thebehaviourofthe model was compared with the situation after the first covid-19 wave in the Czech Republic (March–May 2020) and the post-wave behaviour of the system. We followed the results of own research studies and used primary data focusing on the effects of the covid- 19 pandemic on smes in the Czech Republic, or vis- itorprofilesandsatisfactioninSouthBohemia.More- over, we used a range of studies about the covid-19 impacts on the hospitality and tourism industry pub- lished by the unwto and the Czech Tourism Board (https://www.unwto.organd https://tourdata.cz /temata/data/). Results We simulated three possible scenarios (Scenario 0, Scenario1,andScenario2)offuturetourismdevelop- ment in connection with the hospitality and tourism industry restart in the post-covid-19 period. These three possible future scenarios are confronted with the theoretical situation without the covid-19 dis- ease.Using Vensim 6 Professionalsoftware,we utilise theSyntheSimfunctionforscenariossimulations. Scenariowithoutthe covid-19Disease In this scenario, we simulate the theoretical situation of how the hospitality and tourism industry in South Bohemia would be developing if the covid-19 pan- demic had not occurred. The development would be natural and continuous without any external impacts andspecificpolitics. Scenario0 WeconsiderScenario0asthebasesituationwhenwe considerthebasicimpactsofthe covid-19pandemic. We changedthe input parametersas follows: • 43decreaseoftheNumberofvisitordays based onthestatistics(https://tourdata.cz/temata/data/). • The Human resources competency index de- creasedfrom0.5to0.4becauseoftheprofession- als and employees outflow from the hospitality and tourismindustry. • TheCompetitioninthehospitality&tourismin- dustryindexdecreasedfrom0.8to0.6duetothe closing of businesses as a result of the covid-19 pandemicrestrictions. Scenario1 Scenario1isconsideredasapessimisticsituationwhen peoplewillbegenerallyscaredtotravel.Incomparison with the base situation (Scenario0), we consider30 fewerovernightsandone-day-visitorsin Scenario1. Scenario2 Scenario 2 is considered as an optimistic situation when people will be generally anxious to travel since theywerenotabletogoonholidaysduringthecovid- 19 pandemic. In comparison with the base situation (Scenario 0), we consider 30 more overnights and one-day-visitorsinScenario1. SimulationsResults The simulation results show that the number of visi- torsanddaysspentinSouthBohemiaafterthetourism 130 | Academica Turistica, Year 14,No. 2,December 2021 PetrŠtumpf etal. Restart of Hospitality and Tourism Figure3 ScenarioSimulations:NumberofVisitor-Days restart could drop quite dramatically (Figure 3). If we consider the optimistic Scenario 2, the number of visitor-days will be 72 of the situation without covid-19attheendofthesimulation(step120).How- ever,ifweconsiderthebasesituation(Scenario0)and the pessimistic Scenario 1, the number of visitor days will be 44 of the situation without covid-19 (Sce- nario0),or13respectively(Scenario1),attheendof thesimulation(step 120). Fromthesimulationresults,wecananalysethesit- uation in the hospitality and tourism industry. The simulation shows how Profit&Loss develops in par- ticularsituations.Whiletheaccommodationindustry willachieveprofitsonlyintheoptimisticScenario2at the end of the simulation period (Figure 4), the other hospitality and tourism services will be profitable in the optimistic, as well as in the base, situation at the endof thesimulation (Figure5). Figure 6 shows the development of accommoda- tion establishments occupancy. The results show that thestabilisationoftheaccommodationsectorwilllast significantlylongerin pessimistic Scenario1. The simulated scenarios showed a possible devel- opment of the hospitality and tourism industry in the South Bohemia Region, the Czech Republic. The simulation shows that the recovery after the tourism restartwillnotbeeasy,andthehospitalityandtourism industry will suffer from several related problems, suchastheoutflowofhumanresourcesfromthe h&t sector. DiscussionandConclusion A tourism destination is considered to be a dynamic complex system. Managing tourism destinations is Figure4 ScenarioSimulations:Accommodation EstablishmentsProfit&Loss Figure5 ScenarioSimulations:Other h&t services Profit&Loss Figure6 ScenarioSimulations:Accommodation EstablishmentsOccupancy uncertain,anddestinationmanagershavetomakede- cisions in a complex environment, including many stakeholders with different management objectives and interests (Mai & Smith, 2018). System dynamics in travel and tourism research was used by many re- searchers(Borštnaretal.,2011;JereJakulin,2016;2017; JereLazanski&Kljajic,2006;Mai&Smith,2018;Pat- terson et al., 2004; Ropret et al., 2014; Sedarati et al., 2019;Štumpf&Vojtko,2016;Tanatal.,2017;Tegegne etal.,2018;VojtkoandVolfová,2015).Ourstudyiden- Academica Turistica, Year 14,No. 2,December 2021 |1 31 Petr Štumpfetal. Restart of Hospitality and Tourism tifiesthecomplexityofthedestinationsystemusinga Stocks and Flows Diagram andsimulation modelling. Moreover,weusethemodelforscenariossimulations in the covid-19 tourismcrisis. The proposed system dynamic model can be con- sidered as a unique tool for destination managers to understandanddealwiththesoftsystemsandtourism development policies which determine the dynamics of the destination system. The model enables us to simulate differentcombinations of possible future de- velopment, the effects of decisions and policies, and totesttheireffectivenesstofindtheoptimalsolutions, not onlyin crisis situations.Therefore,theresultscan be used practically by destination managers for des- tination planning and destination strategies formula- tion. Thetheoreticalcontributionofthemodelliesinits complexity, and it covers the crucial relations in the destination system respecting the economic, social, and environmental sustainability of tourism. These facts underline the necessity of modelling the desti- nation system properly to achieve efficient destina- tion management (Bieger, 2008; Farrell & Twining- Ward,2004;Lew&McKercher,2006;Rodriguez-Diaz & Espino-Rodriguez, 2007). The research question was formulated:How will thehospitalityandtourismindustrydevelopinthepost- covid-19eraintheSouthBohemiaRegion? The sim- ulatedscenariosshowthepossibledevelopmentofthe hospitalityandtourismindustryintheSouthBohemia Region, the Czech Republic. The simulation shows thattherecoveryoftourismwilldevelopdifferentlyin various situations, depending on tourist behaviour in the post-covid-19 era (long-lasting fear of travel, on onehand,andatravelboom,ontheotherhand).How- ever, the hospitality and tourism industry will suffer from several related problems, such as the closing of tourism businesses, or outflow of human resources from the h&t sector. BasedonJereLazanskiandKljajic(2006),thepro- posed system dynamic model was established by the authors,astheobservers/descriptorsofthemodel.We canconsiderthisfactasalimitationofthestudyasthe model may be influenced, to a certain extent, by the authors’ perspective. Other limitations of the model areconnectedwiththecalibration.Wehadtoestimate severalvariables’quantificationandtheirinitialvalues basedonexperts’opinions.Moreover,it is noteasyto set the relations between several variables as graph functions since they usually interact in a non-linear way (Baggio & Sainaghi, 2011; Mai & Smith, 2018). Therefore,wewerenotabletovalidatethesimulation results in their absolute values, but the simulations can point to future development and the differences betweenvariousscenarios. The systems approach and complex system dy- namics modelling deserve better attention in future research, in terms of social, environmental, and eco- nomic sustainability in tourism destinations. These methodsrepresentthescientifictoolsthatcanprovide balanced, optimal results to find a consensus among different aims of various stakeholders in tourism des- tinations.The proposed model can be useful for sim- ulationsvarietyscenariosofthedestinationsystemin connectionwithpost-covid-19travelbehaviour.The precise calibration for the situations in a variety of destinations is the way for future research. This crisis oftourismhasshownanenormousandsuddendrop in international travel and the reduction of business activitiesinthehospitalityandtourismindustry. The dynamics of tourism and simulations of the post-covid-19scenariosrepresentabigchallengefor the future. The current situation outlines the neces- sity of a complex and systemic approach in managing tourism destinations. Therefore, we consider our sys- temdynamicsmodelausefultoolfordecision-making support and sustainable destination development in thepost-covid-19era. References Ahlert,G.(2008).Estimatingtheeconomicimpactofanin- creasein inbound tourism on theGerman economy us- ing tsa results. JournalofTravelResearch,47 (2), 225– 234. Andergassen, R., Candela, G., & Figini, P. (2013). An eco- nomic model for tourism destinations: Product sophis- tication and price coordination.TourismManagement, 37, 86–98. Athanasopoulos, G., & Hyndman, R. J. (2008). Modelling and forecasting Australian domestic tourism. Tourism Management,29(1),19–31. 132 | Academica Turistica, Year 14,No. 2,December 2021 PetrŠtumpf etal. Restart of Hospitality and Tourism Baggio, R., & Sainaghi, R. (2011). Complex and chaotic tourism systems: Towards a quantitative approach.In- ternational Journal of Contemporary Hospitality Man- agement,23(6),840–861. Bieger,T.(2008).ManagementvonDestinationen.DeGruy- terOldenbourg. Bonham, C., Gangnes, B., & Zhou, T. (2009). Modeling tourism: A fully identified vecm approach. Interna- tionalJournalofForecasting,25(3),531–549. Borštnar,M.K.,Kljajić, M.,Škraba,A.,Kofjač, D.,&Rajko- vič, V. (2011). The relevance of facilitation in group de- cision making supported by a simulation model.System DynamicsReview,27(3),270–293. Buchta, C., & Dolnicar, S. (2003). Learning by simulation: Computer simulations for strategic marketing decision support intourism.InternationalJournalofTourismSci- ences,3(1).https://doi.org/10.1080/15980634.2003 .11434540 Farrell, B. H., & Twining-Ward, L. (2004). Reconceptualiz- ingtourism.AnnalsofTourismResearch,31(2),274–295. Forrester, J. W. (1961). Industrial dynamics. Productivity Press. Garson,G.D.(2008).Computerizedsimulationinthesocial sciences:Asurveyandevaluation.Simulation&Gaming, 40(2),267–279. Greiner,R. (2010). Improving the net benefitsfrom tourism for people living in remote Northern Australia.Sustain- ability,2(7),2197–2218. Jackson, M. C. (2003).Systemsthinking:Creativeholismfor managers.JohnWiley& Sons. JereJakulin,T.(2016).Systemdynamicsmodelsasdecision- makingtoolsin agritourism.Agricultura,13(1–2),5–10. JereJakulin,T.(2017).Systemsapproachtotourism:AMeth- odology for defining complex tourism system.Organi- zacija,50(3),208–215. JereJakulin,T.(2019).Systemsapproachtoculturaltourism and events.AcademicaTuristica,12(2),185–191. Jere Lazanski, T., & Kljajic, M. (2006). Systems approach to complex systems modelling with special regards to tourism.Kybernetes,35(7/8),1048–1058. Kaspar,C.(1976).Letourisme,objetd’étudescientifique.The TouristReview,31(4),2–5. L a c i t i g n o l a ,D . ,P e t r o s i l l o ,I . ,C a t a l d i ,M . ,&Z u r l i n i ,G . (2007). Modelling socio-ecological tourism-based sys- tems for sustainability. Ecological Modelling, 206(1–2), 191–204. Laesser,C.,&Beritelli,P.(2013).St.Gallenconsensusondes- tination management.JournalofDestinationMarketing &Management,2(1),46–49. Lawson, S. (2006). Computer simulation as a tool for plan- ningandmanagementofvisitoruseinprotectednatural areas.JournalofSustainableTourism,14(6),600–617. Lew, A., & McKercher, B. (2006). Modeling tourist move- ments.AnnalsofTourismResearch,33(2),403–423. Liu, C.-H., Tzeng, G.-H., & Lee, M.-H. (2012). Improving tourism policy implementation – The use of hybrid mcdm models.TourismManagement,33(2),413–426. Mai, T., & Smith, C. (2018). Scenario-based planning for tourism development using system dynamic modelling: A case study of Cat Ba Island, Vietnam.Tourism Man- agement,68, 336–354. Patterson,T.,Gulden,T.,Cousins,K.,&Kraev,E.(2004).In- tegrating environmental, social and economic systems: A dynamic model of tourism in Dominica.Ecological Modelling,175(2),121–136. Ritchie, J. R. B. (2003).Thecompetitivedestination:Asus- tainabletourismperspective. cabi. Rodriguez-Diaz,M.,&Espino-Rodriguez,T .F .(2007).A model of strategic evaluation of a tourism destination based on internal and relational capabilities.Journalof TravelResearch,46(4),368–380. Ropret, M., Jere Jakulin, T., & Likar, B. (2014). The systems approachtotheimprovementofinnovationinSlovenian tourism.Kybernetes,43(3/4),427–444. Schianetz, K., Kavanagh, L., & Lockington, D. (2007). The learning tourism destination: The potential of a learn- ingorganisationapproachforimprovingthesustainabil- ityoftourismdestinations.TourismManagement,28(6), 1485–1496. Sedarati,P .,Santos,S.,&Pintassilgo,P .(2019).Systemdy- namics in tourism planning and development.Tourism Planning&Development,16(3),256–280. Senge, P. M. (1990).Thefifthdiscipline:Theartandpractice ofthelearningorganization (1sted).Doubleday. Sterman, J. (2000).Businessdynamics:Systemsthinkingand modelingforacomplexworld. McGraw-Hill Education. Štumpf, P., &Vojtko,V. (2016).The systemdynamicsmodel forsupportofthedestinationmanagementinSouthBo- hemia.BusinessTrends,6(4),43–61. Tan, W.-J., Yang, C.-F., Château, P.-A., Lee, M.-T., & Chang, Y.-C. (2018). Integrated coastal-zone management for sustainable tourism using a decision support system basedonsystemdynamics:AcasestudyofCijin,Kaohsi- ung,Taiwan.Ocean&CoastalManagement,153,131–139. Tegegne,W .A.,Moyle,B.D.,&Becken,S.(2018).Aqualita- tive system dynamics approach to understanding desti- nation image.JournalofDestinationMarketing&Man- agement,8,14–22. Academica Turistica, Year 14,No. 2,December 2021 |1 33 Petr Štumpfetal. Restart of Hospitality and Tourism Vojtko,V.,&Volfová,H.(2015).Regionalsustainabletourism – A system dynamic perspective. In L. Novacká & G. Ivankovič (Eds.),Tourism&hospitality–sustainability andresponsibility(pp. 21–40).Profess Consulting. Appendix1:DetaileddescriptionoftheStocksandFlows Diagramstructure 1. Accommodationestablishments(ae)capacity represents oneofthekeystockvariablesintheentiremodel,which is expressed by the number of beds in the destination. The capacity of accommodation establishments (ae) is increased byinvestments (flow variable; inflow). The to- tal investments in the construction of new accommo- dation capacities (beds) include either investments due to the need to build capacities (aebuilding necessity) – to extend the capacity of the existing aes, or addi- tionalinvestments, in other words, construction of new aes. In this case, the additionalinvestments represent an exogenous variable. In general, however, they may be determined, for example, by the attractivenessof the tourism sector in the destination. It can be increased, for example, by subsidies for the construction of new accommodation capacities.Theaebuildingnecessity is given by the occupancy of accommodation facilities, which is expressed as the ratio between the number of overnight stays (per month) and the capacity of accom- modation facilities (per month). The total capacity of aes is reducedbytwo flow variables– a depreciation of accommodationfacilities(outflow)andaeclosing (out- flow). While the depreciation is mainly caused by the occupancy of accommodation facilities (the higher oc- cupancy of the accommodation facility, the higher the wear and tear), the closure of accommodation facilities depends mainly on the monthly financial result of aes and profitability of aes, which in this case is expressed by profitability based onthereturnofsales (ros). If the aes donotreachatleasttheexpected aes’minimaltar- getros, theaccommodationfacilitieswillbecloseddue to their unprofitability. 2. Accommodation services quality is determined by the change in the quality of accommodation services (qual- itychange, outflow variable), which is influenced by ex- ogenous variables – human resources competency and competitioninh&tindustry –asinthecaseofthequal- ity of other services. Furthermore, however, the qual- ity of accommodation facilities is increased by invest- ments placed in accommodation capacities, provided that investments in accommodation facilities exceed their depreciation. The quality of accommodation ser- vicesandthequalityofotherservicesisthenexpressed bytheauxiliaryvariableh&tservicesquality, whichin- fluences, together with other factors, the visitors’satis- factionchange. 3. Otherh&tservicesquality isdeterminedbyOtherh&t servicesqualitychange (flowvariable),whichinthesug- gested model is influenced by two exogenous variables –humanresourcescompetency andcompetitioninh&t industry. In general, it can be assumed that increase in employees’ competencies will increase the quality of, for example, catering, guide, transport, and other ser- vices;similarly,theincreaseincompetitionshouldforce providers to be more competitive and increase the qual- ity. 4. Accommodation real price level is determined by the price level change (flow variable), which is influenced mainlybythe ae occupancy(the effectofaeoccupancy onpricelevel).Ingeneral,itcanbeconcludedthatifthe ae occupancy increases, the price of accommodation willalsoincrease.The price leveliscalculatedasthe av- erage price per bed/night. For simplification, the price level was calculated only for accommodation services. When summarising other services into one common category,quantifyingthepricelevelforallotherservices wouldrequiretheirdetailedelaborationandcalculation in aseparatemodel. 5. Accumulatedinflation in the proposed model represents astockvariablethatneedstobequantifiedduetothe fact that the model considers the real price level for ac- commodation. The inflow of accumulated inflation is monthly inflation(annualinflationrate calculatedfor 12 months of the year with respect to the time unit of the simulation, which is one month). 6. AccumulatedProfit&Lossofh&tindustry is stipulated by monthly Profit&Loss, which represent a flow quan- tityforthepurposeofthismodel.Financialresultsofac- commodationfacilitiesandthefacilitiesprovidingother tourism services (to keep the model as simple as pos- sible the other services were not further distinguished) are reflected in aProfit&Loss(month). Thus, monthly Profit&Loss is calculated using the difference between revenues from accommodation, and fixed plus variable costsofaccommodationfacilitiesandtheexpectedprof- itabilityoffacilitiesprovidingotherservices.Inthiscase, itisexpressedbytheaveragerosofotherh&tservices. 7. Profit&Lossofh&tindustry(year) hadtobequantified not only for monitoring the annual Profit&Loss in the tourismsectorbutalsoforthesubsequentquantification of tax revenues from the h&t industry in the destina- 134 | Academica Turistica, Year14,No. 2,December 2021 PetrŠtumpf etal. Restart of Hospitality and Tourism tion,whicharegeneratedinindividualyears.Theinflow ofProfit&Lossofh&tindustry(year) isalsoProfit&Loss (month) (flow variable). TheProfit&Loss(month) is not accumulated for the whole simulation period as in the previous case, but the Profit&Loss is nullified after each year. This represents an outflow of Profit&Loss of h&t industry (year), and it is possible to derive tax revenue (tr)of h&t industryfrom it. 8. Tax revenues (tr) of h&t industry (year) are in the model (again,withrespect tothetime unitof thesimu- lation) given by the inflow oftaxrevenues(tr)ofh&t industryaccordingtoindividualmonths(flowvariable). Forsimplification,taxrevenuesincludeonlyincometax and vat, which are calculated from the total financial result of accommodation and other tourism facilities,in other words, from the revenues from accommodation and other services. Tax revenues are reduced by the as- sumedgreyeconomyratio(exogenousvariable).Follow- ing each year, tax revenues are nullified (flow variable). It is an outflow of annual trs of the h&t industry in the destination.It is possible to derive from it in a sim- plified way thetaxrevenuesreturnedbackindestination (flow variable), which is redistributed and returned to thelocalandregionalbudget. 9. trreturnedbackindestination(year) represent a stock variable that has an inflow in the proposed model in theformofthetaxrevenuesreturnedbackindestination (flow variable) and outflow in the form oftrreturned backnullifying in order to determine tax revenues each year.Thisisthewayinwhichthefinancialresourcesare expressed; after the taxes are redistributed the financial resourcesreturntothedestinationthroughlocalandre- gional public budgets. Their share of the total tax rev- enuesfromthetourismsectorinthedestinationwillde- terminethebudgetallocationoftaxes.Thisfacthasbeen simplified for the purpose of thismodel to a singlecoef- ficient oftrreturnedbackindestinationratio (exoge- nousvariable).AnotherfactoristheLocalbusinessesra- tio based in the destination (exogenous variable). Busi- nessentitieslocatedoutsidethedestination,whichpro- videtourismservicesintheregion,filetheirtaxreturnat the place of their registered office. This fact reduces the taxrevenuesthatflowbacktothedestination. 10. Visitor-days per year (Visitor-days 12m) is a key stock variable on the demand side. The inflow is (with re- spect to the time unit of the simulation), for the pur- poses of this model, expressed inanumberofvisitor- dayspermonth which is given by the sum ofovernights andone-dayvisitors. In the proposed model, the num- ber of one-day visits is influenced by ‘word-of-mouth’ (effectofwomonone-dayvisitors), individual market- ing communication of other service providers (effectof imc/otherh&tservices/onone-dayvisitors ), andother effectsonone-dayvisitors(exogenousvariable). The number of overnights can be increased through higher expenditures that the accommodation facilities spendonmarketingcommunication(effectofimc/ae/on overnights), but also by providers of other services (ef- fect ofimc/otherh&tservices/on one-day visitors ). A wider offer of other services or higher awareness of the offer can encourage visitors to stay longer. The num- ber of overnights will be further increased by higher expendituresonmarketingcommunicationinthedesti- nation(Destinationmc), more intensivepositive ‘word- of-mouth advertising’(effectofwomonovernights),de- clining price level (effectofpricelevelonovernights, and related exchangerateeffect as an exogenous variable), orothereffectsonovernights. The number of overnight staysisalsoinfluencedbytheaveragelengthofstaytrend (asanexogenousvariable),whichisbasedontheglobal trendofshorteningthelengthofstayof tourismpartic- ipants in destinations. This fact is due to the preference of tourism participantsto travelseveral timesa yearfor shorterstays.Thenumberofovernightstaysisthenlim- ited by the capacity of accommodation facilities. Exogenous variables affecting the number of visitor- days (other effectsonone-dayvisitors andother effects onovernights)wereusedasaninputvariableforsimu- lation of future development scenarios for the restart of the tourism sector after the covid-19 era. 11. Visitor-daysinthelast24months (Visitor-days 24m) is a stock variable which, in the proposed model, has the same inflow asVisitor-days12m, and which is the basis forquantifyingthewompotential.Theproposedmodel assumesthatvisitorswhohavevisitedthedestinationin thelast24monthswillsharetheirexperiencewithother possiblevisitorstothedestination(theirrelativesandac- quaintances).This means that the visitor-days from the previous 24 months can generate more visitor-days in the future. However, only satisfied visitors will share a positiveexperience. 12. Visitors’ satisfaction is determined by the satisfaction change(flowvariable),whichinthismodelisinfluenced by theh&tservicesquality of services, the state of the culturalandnaturalpotential(cnp), the real price level oftheaccommodation services,andtheleveloftheRes- identsirritation from tourism. In general, it can be con- cluded that the satisfaction of visitors will grow in line Academica Turistica, Year14,No. 2,December 2021 |1 3 5 Petr Štumpfetal. Restart of Hospitality and Tourism withbetterculturalandnaturalpotential,inotherwords with better primary attractiveness of the destination, if thepricesdecline,but thequalityofservicesgrows,and thelocalswillbemorefriendlytovisitors.Thesatisfac- tion of visitors is expressed on the scale in the interval of [0,1]. The value of 0 means that visitors to the desti- nation are completely dissatisfied;in contrast, the value of1isassumedinasituationwhere thevisitors would be entirelysatisfiedwiththeirstayinthedestination. 13. Residents’irritation,intheproposedmodel,isinfluenced mainly by thetourismintensity,which in thiscase is ex- pressed by the ratio of the number of visitor days per month to the number of local inhabitants. The second influence thatisreflectedintheirritationof residentsis the cultural and natural potential. Theeffectoftourism intensityonirritation and theeffectofculturalandnat- uralcapacityonirritation results inthechangeinirri- tationoflocalinhabitants, which represents a flow vari- able affecting the current stateoftheresidents’irritation oflocalinhabitants.Ingeneral,inthisrelation,itcanbe concludedthattheincreasingintensityoftourisminthe destination increases the irritation of the local popula- tion, while the improving cultural and natural environ- mentreducestheirritationofresidents.Theirritationof local people is expressed on the scale in the interval of [0,1]. The value of 0 means that the local people in the destinationarenotirritatedbythepresenceofvisitorsin thedestination.Incontrast,thevalueof1isassumedina situationwherelocalswouldbeupsetaboutthepresence of visitors and the negative consequences of tourism as much as possible. 14. Cultural and natural potential (cnp) in the proposed model is mainly influenced by the number of visitor- days.Ingeneral,itmaybeconcludedthatthemoredays tourists and visitors spend in the destination, the more they will burden the naturalenvironment and affectthe localculture,thusdegradingtheprimarycapacityofthe destination. Theeffectofvisitor-daysoncnp andother effectsoncnp (exogenous variable) results in thecnp change, which is a flow variable affecting the current state ofcnp. Other impacts on the cnp can be invest- ments in historic preservation, environment protection, or generally in improving the attractiveness of the pri- mary capacity of the destination. The model also takes into account a certain degree of self-renewal, especially of the natural capacity of the destination. In this case, favourable conditions for the self-renewal of the des- tination are assumed, such as an appropriate environ- mental protection policy or prevention of ‘brownfields’ creation. cnp is expressed on the scale in the interval of [0,1]. The value of 0 assumes a borderline situation wheretherewouldbenonaturalandculturalcapacity in the destination which was creating an attractiveness for tourism. In contrast, the value of 1 is assumed in a situation where the natural and cultural capacity of the destinationisatthehighestpossible level. 136 | Academica Turistica, Year14,No. 2,December 2021