Study The Spatial And Temporal Characteristics Of Space Of Flows

With the support of information and communication technology?ICTs?,production and organization are no longer limited to fixed geographical locations.Function and power are organized in flows.The dominant spatial form is no longer“space of places?SPs?”,but a new kind of space named“space of flows?SFs?”.The SFs breaks through the limit of time and space and guides the development of future regional space pattern to some extent.It has been extending to the entire field of human activities and their interiors through the interaction of flows.The population flow,the material flow,the capital flow and the information flow are the main elements of SFs.Among these elements,the information flow is playing a leading role.The non-material flow can affect the space directly by attaching to the information flow.With the help of informatization and intelligentialization,the flow of elements can be configured and organized more and more efficiently by the material flows,and then the spatial pattern would be affected.However,since the data related to information flow often involve privacy information making it is difficult to obtain.Therefore,the current research on factor flows is more focused on the population flow and the material flow.After studying the theory of SFs,this paper takes into account the availability of data and then taking ETC data of Taiwan Expressway in 2017 as fundamental data to study the temporal and spatial differences of population flow and material flow in Taiwan Province.The research contents mainly include:1)the formation and development of the concept of SFs and the association of SFs and SPs;2)the theory of SFs and its interaction:a)the embodiment of the three elements of space in the SFs;b)the definitions and functions of the material flow,population flow,information flow and cash flow etc.and the relationship between each other;c)the labor division theory,interdependence theory and competitive cooperation theory are introduced to analyze the interaction mechanism of SFs;3)the study on temporal and spatial characteristics of population flow and material flow in Taiwan Province;4)the research on population flow and material flow networks of major counties and cities in Taiwan Province;and 5)the prediction of highway traffic?people flow and material flow?in Taiwan Province.The technologies,models and methods used in this paper mainly include spatial-temporal autocorrelation analysis,social network analysis,and time series analysis and so on.The main conclusions are as follows:?1?SFs depend on SPs,but beyond SPs,and the two spaces will coexist for a long time.The impact of the SFs on the SPs is increasing.SPs could exercise its district function depends on the organization of SFs.However,the SFs have not completely penetrated into all areas of human social practice.With the help of the instant flow of information,SFs can break through the limit of time and geographical distance to a great extent,and cause the reduction of the influence of geographic distance.?2?In the SFs,each participant has its own“position”or“location”.There are“distances”between the participants.Elements of the SFs flow caused by the demand,attraction or other external force,and to form a kind of inherent association space.The traditional three elements of space are also reflected in the SFs.?3?All kinds of elements in the SFs interweave and influence each other and promote the development and evolution of the SFs.Any element is indispensable,while the information flow is at the core position and plays a guiding role in the formation and development of the SFs.?4?The research results of the temporal and spatial characteristics of population flow and material flow in Taiwan province based on ETC data of Expressway show that:1)the time differentiation of the“double-flow”is obvious,which shows that there is a large gap in the volume of the“double-flow”in different months,days and periods.Under different research units,the peak periods of different types of flows are staggered relatively.2)The spatial differentiation of the“double-flow”is also obvious,which shows that the hot spots of different types of traffic are different.What's more,the total volume of the“double-flow”of different counties and cities is significantly different.3)Compared with material flow,the contribution of population flow to the total traffic of Taiwan's expressway network is much greater than that of the former.4)There is a significant positive spatial-temporal correlation in Taiwan's Expressway network,but this correlation is mainly reflected between the first-order spatial adjacent sections.?5?Four different associative networks?population flow associative network in county-city-level and station-level,material flow associative network in county-city-level and station-level?are constructed to study the spatial associations among the major counties and cities in Taiwan Province.The results show that:1)Without considering the difference in the volume of the population flow and material flow among the research units,there are population and material links between individual research units,indicating that the space of population flow and material flow in Taiwan Province has entered a period of network homogeneity.2)The overall densities of both population linkage network and material linkage network in Taiwan Province are low.The flows of population and material are more likely to be carried out between neighboring counties and cities.3)According to the analysis of the restriction index in the structural hole,Tainan City,Taichung City,Taipei City,Xinbei City and Xinzhu County are in the dominant position in the competition with other counties and cities in the network in terms of population linkage network while Kaohsiung City,Jiayi County,Taoyuan City,Xinbei City,Xinzhu City,Xinzhu County and Changhua County are are in the dominant position in terms of material linkage network.4)As far as individuals are concerned,Xinbei City is not only the core distribution center of population flow,but also the core distribution center of material flow.Taichung City is the hub node of population and material linkage networks,which plays the function of connecting different subgroups.It is also the transit station of population flow and material flow.?6?In order to find more suitable traffic volume prediction models,the selection of prediction models in three different situations is studied.The results show that:1)the time series of original population flow and material flow in Taiwan Province are not stable and seasonal?periodic?with a period of seven days.The seasonal ARIMA model can be used for forecasting and analyzing the traffic volume.2)In three cases,the suitable prediction models for population flow prediction are respectivelyARIMA?2,1,1??3,1,1?₇,ARIMA?4,1,1??3,1,1?₇ andARIMA?4,1,1??3,1,1?₇.The differences among the three models are only that the values of parameters p are slightly different.Overall,it is suggested to use the prediction modelARIMA?4,1,1??3,1,1?₇ as Taiwan's population flow prediction model.3)In three cases,the suitable prediction models for materialflowpredictionarerespectivelyARIMA?2,1,2??1,1,2?₇,ARIMA?0,1,2??1,1,2?₇ andARIMA?5,1,2??1,1,1?₇.The differences among the three models are that the values of parameters p and Q are slightly different.Overall,it is suggested to use the prediction modelARIMA??0,2,5?,1,2??1,1,2?)₇ as Taiwan's material flow prediction model.