Improving cycling environment in a Green Park based on the post-occupancy evaluation method

ABSTRACT Nowadays, urban cycling activities in China are developing rapidly, and the number of cyclists has also increased dramatically. The improvement of urban cycling environment design in cities has positive significance for improving the quality of living environment. However, the related research about the environmental design theory is relatively scarce in China. This paper tries to apply the Post-Occupancy Evaluation (POE) method to evaluate the cycling environment in a green park. This study investigates the problems of the cycling environment of a park in Qingdao and then proposes some methods for improvement. The evaluation model is established by using the analytic hierarchy process (AHP) to assess the cycling environment. POE has been carried out by collecting data through questionnaires. The low score of POE has shown the shortcomings of the green park cycling design. Furthermore, this study proposes some principles of systematization, people-oriented, refinement and ecological economy to optimize the cycling environment of the green park. This paper creates a new method of designing the cycling environment, namely “POE – behavior analysis – design optimization”, which provides a theoretical reference for cycling environment design of the green park. GRAPHICAL ABSTRACT


Introduction
With the progressing construction of the park, more and more cities have begun to integrate existing detached park resources with the surrounding area and are also determined to construct of the urban park in the local area. The green path influences and promotes cycling that has now become a challenge for sustainable urban mobility in many cities where this mode of transport carries little weight (Campos-Sánchez, Valenzuela-Montes, and Abarca-Álvarez 2019). The indirect emission intensity effect was a leading contributor to change of carbon emission (2017). Greenway, developed from the Parkway, has become the focus of provincial and urban construction after 2005. In the last a few years, cycling mobility has been growing rapidly (Di Mascio et al. 2017;Ritchie 1998). Cycling activities have been used parks or other public places as active venues, and bicycles are used for commuting and leisure time activities (Bíl, Bílová, and Kubeček 2012;Han, Meng, and Kim 2017). Leisure cycling has the characteristics of popularity and adaptability, and it has unique advantages in outdoor sport.
First of all, many scholars studied the historical developments, classification systems and standards of parks, involving attributes, contents and functions of parking space. Abdullah Akpinar (Akpinar 2016) comprehensively evaluated the perception and use of metropolitan green space in Turkey and the factors affecting its use. Giulio Senes conducted a study to confirm the significant correlation between green space use, social population, landscape characteristics, accessibility and time variables (Senes et al. 2017;Liu et al. 2018). In the field of landscape ecological design and open space design, some scholars paid specific attention to cycling environmental design. In 1990, Charles Little wrote Greenways for America that examined the recreational function, walking and cycling system as an organic part of the Greenway system. Greenways, blue ways, skyways and other ways to a better London was published by Tom Turner in 1995. He summarized the green ways and summarized six types of green way containing parkway and cycleway from the angle of the functional requirements (Turner 1995). Some scholars showed the link between urban green planning and environment-friendly mobility policy, and encouraged green streets as a multifunctional part of the urban green infrastructure (Nawrath, Kowarik, and Fischer 2019;Zhao, Gao, and Knoop 2019;Cho, Huang, and Huang 2018).
On this basis, many studies confirmed the promoting effect of "trail" environmental design of cycling activities and the reverse effect of cycling behavior on ecological environment (Clay and Smidt 2004;Wegman, Zhang, and Dijkstra 2012). By analyzing practical cases, some studies analyzed the usage patterns and respective demands of different types of cyclists and leisure groups on the Greenway from the perspective of demand. The social elements in the trail design were studied, including the park manager's attitude towards mountain bikes and the interaction of community participation in the environmental construction of the park (Park et al. 2010). Meanwhile, there is a case study of the parkway represented by Greenway from Brooking to Queens, New York (Jacobsen 2015). A lot of research on how to analyze whether the current road provides a suitable model for cycling. They involve the cycling environment, road lighting design, weather factors, visibility, and summing up the causes of daily fluctuations in cycling behaviors (Meitner 2004;Ryan, Fábos, and Allan 2006); Some scholars analyzed the cycling environment design in New York, Hague and Toronto (Jungnickel and Aldred 2013;Shannon, Smardon, and Knudson 1995;Mehaffy 2008). Through interviews from five European countries (Netherlands, Belgium, Germany, the United Kingdom and Denmark), some scholars discussed the definition of bicycle highway and the influence of their conceptualization of bicycling experience on the physical design of bicycle road (Liu et al. 2019;Mohamed and Bigazzi 2019).
Compared to other countries, China has unique bicycle routes such as the Great Wall in Beijing, city bicycle tours in Shanghai and Suzhou, and adventure tours in Lhasa, Tibet and Sichuan province (Bikechina 2016;Chinacycletours 2016). At the same time, the Chinese government has been attempting to stimulate the development of a bicycle-related infrastructure in order to build a well-designed bicycle network (Karki and Tao 2016). Bike-sharing systems have been growing at an unprecedented pace: almost doubling in less than 5 years to nearly 700 systems (Fishman, Washington, and Haworth 2014).
This paper applies post-occupancy evaluation (POE) to the field of cycling environment design, and innovatively proposes the design method of "POE -behavior research -design optimization". This paper summarizes the contents of the urban cycling environment and deepens the results of a POE survey through behavior research. This paper discusses how to optimize urban cycling environment based on park reconstruction, and optimizes the design of the various contents of the urban cycling environment, so as to provide references for relevant urban environmental design practices.

Research method
Environmental impact assessment of design has received significant attention in the recent years (Zhu and Deshmukh 2003). Firstly, through the method of literature reading and analytic hierarchy process (AHP), this paper evaluates the cycling environment of Green Park, and sums up the cycling environment into four contents: "Cycling road system", "Landscape environment system", "Service facility system" and "Night cycling environment". From the point of view of behavioral psychology, the comparative study and mathematical statistics are utilized to analyze the cycling behavior, and the difference between cyclist and noncyclist in behavior law is analyzed. Secondly, based on the shortcomings of the evaluation and the behavior law found in the questionnaire, this paper sums up the principles that should be followed in the design of urban park cycling environment, analyzes the corresponding design elements and optimizes them. Finally, focusing on the urban cycling environment, this paper summarizes the above main conclusions, and puts the simultaneous design research method of postoccupancy evaluation and behavior analysis to a certain extent. The design should follow the principles of systematization, refinement and fine ecology. The urban cycling environment design based on postoccupancy evaluation can be closer to the life of users and meet their demand.

Analytic hierarchy process
Analytic Hierarchy Process (AHP) is divided into three steps: establishing hierarchical structure, determining factor weights and evaluating scores.

Establishing a hierarchy
According to the relevant research results, the design elements are selected, with no more than 7 at each level.

Determinant weight
By using 1-9 scale method and comparing with expert matrix questionnaires, the relative importance score of each evaluation factor is obtained as the original parameter, which can be calculated by substituting the comparison matrix (Formula 2.1). The weight values of each factor are obtained, and the weight values obtained need to pass the consistency test (c i < 0.1, formula 2.2). 1)

Assessment and scoring
By using the scaling method to collect the original data, it will be converted into the index grade score. Then, according to the component evaluation matrix (formula 2.3) and using the principle of fuzzy mathematics (formula 2.4), the scores of various evaluation factors can be obtained from bottom to top.

4)
Ri ¼ In Formula 2.4, "B i " denotes the evaluation result of U i (standard level evaluation factor u i ); "A i " represents a fuzzy vector consisting of the weights of each secondary evaluation factor u i in the i-th secondary factor set U i ; "R i " represents the fuzzy evaluation matrix of the corresponding secondary evaluation set U i ; "°" stands for compound operation. The common types are "weighted average operator" and "Zadeh Operator", and the former is selected by this study according to actual needs.

Establishing hierarchical structure
This paper divides the components of the Green park's cycling environment into three levels: the target level, the criterion level, and the index level. The first-level evaluation factors (target level) constitute a set of factors U = {u 1 , u 2 , . . ., u i , . . ., u n }, and each factor is determined by a different secondary evaluation factor (indicator level), that is, U i = {u i1 , u i2 , . . ., u ij , . . ., u im }.

Determining first-level (Criterion level) evaluation factors.
As shown in Figure 1, the classification criterion on the left takes into account the needs of various Greenways and has high authority and objectivity, but field observations show that there are more nighttime cycling activities in the green park. Therefore, according to the actual situation of Green Park, this paper adjusts as follows: (1) Combining "Walking and cycling system" and "Bicycle and car system" into "Cycling road system" by focusing on the evaluation of cycling road itself; (2) Replacing "Green corridor system" with "Landscape environment system" by focusing on the evaluation of green park landscape effects. (3) Merging "Service facility system" and "Information marking system" into "Service facility system" by focusing on the evaluation of ancillary facilities and service facilities.

Determining secondary (Index level) evaluation factors.
1. According to the following steps, the evaluation factors in the "Cycling road system" are screened: (1) With reference to the research in the field of Greenway, "layout selection" is relatively independent and is the basis of road design. It mainly includes: route layout, direction, length, density, quantity, and level setting. It also includes both macro-level resource utilization and micro-level spatial effects and advantages and disadvantages.
(2) Many researches of road landscape emphasize "spatial form", which mainly includes: road alignment, pavement composition, section form, road paving, and spatial interface. In the field of Greenway research, these contents mostly appear in the form of "measurable indicators" and "technical standards".
(3) "Interesting" reflects the attractiveness of cycling routes. At the same time, according to our preliminary research, cycling safety is one of the important factors preventing the public from participating in cycling activities, and "safety" also plays an important role. Therefore, "safety" and "interestingness" are important indicators.
(4) "Park and Ride" to evaluate the degree of connection between park cycling environment and urban road system.
Finally, the secondary evaluation factors of "Cycling road system" are determined as follows: U 1 = {u 11 (route selection layout), u 12 (space form), u 13 (parking and transfer), u 14 (security), u 15 (interesting)} 2. According to the following steps, the evaluation factors of "Landscape environment system" are selected: (1) The traditional landscape research mainly classifies landscape from the static point of view. It mainly includes (micro) topographic landscape, animal and plant landscape, natural water landscape, seasonal meteorological landscape, architectural sketch landscape and so on. According to planning direction, design content, construction results and utilization of Tangdaowan Park, four static indicators, topography and landform, garden plants, landscape waters, landscape sketch, were selected in this study.
(2) Research on the evaluation of road landscape space, urban road landscape and landscape road landscape emphasizes the dynamics of road landscape. That is to emphasize the continuous visual effect of road landscape (shown in Table 1) Finally, the secondary evaluation factors of "Landscape environment system" as follows: U 2 = {u 21 (topography), u 22 (landscape plant), u 23 (landscape water body), u 24 (landscape sketch), u 25 (vividness), u 26 (harmony)} 3. In the Greenway system, the service facility system mainly includes management facilities, entertainment facilities, commercial service facilities, safety and security facilities, popular science education facilities, environmental sanitation facilities and so on.
Tangdaowan Park is of moderate scale. Only recreational facilities and commercial service facilities are closely related to cyclists. The former is mainly embodied in the form of leisure and service stations, while the latter is mainly combined with the open space of the park. The visibility and identifiability of information and signs in the cycling environment of the park are assessed through the "Information marking system".

Determining the weight of each factor
For the cyclist environment of urban parks, different design elements have different level of importance. Therefore, each evaluation factor (u i ) has only one weight value(w i )to indicate its importance, and the greater influence it has, the bigger the impact of weight value is, and vice versa.
Matrix questionnaires were used in this study, and they were compared with each other. Through the 1-9 scale method, experts consulted on the relative importance of evaluation indicators. The experts consulted 12 experts, including 5 professors from the planning department, 3 professors from landscape architecture, 1 staff member of government planning department and 3 graduate students from planning, landscape architecture and landscape architecture.
There are two ways to process computational data: root method (geometric averaging method) and the summation method (normalized column averaging method). According to the actual situation, the standard column averaging method was chosen in this study. The weight of each factor can be obtained by calculating the evaluation matrix (formula 2.1) as shown in Table 2. The data passed the consistency test.

Assessment and evaluation
In this study, "Level 5 Richter Scale" was used to collect the original data of the Post Occupancy Evaluation. The "very satisfied" positive attitude was 5, and the "very unsatisfactory" score was 1. In coordination with this, the evaluation results of each evaluation factor were divided into five grades according to the score: The corresponding relationship between each evaluation grade and the score interval is shown in Table 3.

Obtaining evaluation results
Each question multiplies the "proportion of respondents who selected each option" by "item score", and then adds up to get the item score. For example, 5%     choosing "very disagree", 10% choosing "relatively oppose", 20% choosing "fair", 30% choosing "more satisfied", and 35% choosing "very satisfied", the score of the first item "route layout" is "1 * 5% + 2 * 10% + 3 * 20% + 4 * 30% + 5 * 35% = 3.8 (points)" in the "3.5 ~ 4.5" interval, and it belongs to "good". The final score is shown in Table 4. Through summarizing the results, we can get the following results: "Safety and convenience" is the most prominent problem in the green park cycling environment at present. The problems mainly focus on the following aspects: route selection and space-form construction, service facilities and information marking system design optimization, night cycling safety construction.

Qingdao Tangdaowan Park
As shown in Figure 3, Tangdaowan Park is located in the center of the coastal zone on the south side of Qingdao West Coast Economic Development Zone. In the future, Qingdao will also use cycling roads and other cycling activities to develop cycling events. At present, the West Coast New District has planned four green path cycling routes. Tangdaowan Park covers an area of about 50 hectares, and the coastline is more than 30 kilometers. It is completely open to the public and serves the entire West Coast New Town. The park was renovated in 2012 and has now become a citylevel, multi comprehensive green park. Various bicycles recreational facilities have been built in the park (Figure 4).Tourists intent to use urban bikesharing for cycling on holidays (Kaplan et al. 2015).
The cycling roads of Tangdaowan Park have been carefully designed. The South Coast Park has a coastal walkway on the north side of the coast, and a cycling green road is set in the middle of the park to form a closed cycle route. The ring road serves as the skeleton of the South Bank; The North Shore Park uses the coastal walking trail as a bicycle path and the South Coast coastal walking trail forms the main cycling route that runs through the entire park. In the upper level planning, the coastal walking trail will also form a citylevel cycling route along with the West Ring Road, East Ring Road, and Coastal Avenue. A series of slender, curved, intertwined trail links are connected between the main trunk lines to form a trail network. Therefore, the park forms a cycling road system presenting an overall linear layout and a partial circular layout. Among them, the coastal walking trails and cycling green paths connect different functional areas into a unified whole ( Figure 5).
Tangdaowan Park is close to the district government, and there are many attractions, open spaces, schools, and residential areas within a radius of 10 kilometers. There are more than 20,000 residents living near the park and the population composition is diverse. Therefore, the park was selected as the pilot area of Qingdao's first batch of "public bicycle rental systems". The park and the nearby Coastal Avenue and (East, West) roundabout bicycle lanes were the earliest lines in Qingdao Greenway. Now, it is one of the most frequent urban spaces for cycling in Huangdao District. Therefore, we take Tangdaowan Park as an example to evaluate its cycling environment using POE and AHP methods.

Study on cycling behavior of cyclists in the Tangdaowan Park
According to the theory of "behavior-space", the space is closely related to the behavior of users. Successful design needs to conform to the characteristics of human behavior (such as movement rule, environmental perception, behavior psychology, etc.). By focusing the evaluation of some low sub-items, this study investigates the population composition, behavior rule, environmental perception of the Tangdaowan Park cyclists, and further explores its deep-seated reasons for subsequent design optimization. First of all, a questionnaire survey is carried out to conduct a comparative study of "cyclists" and "noncyclists" to analyze the population composition and behavior rules of cyclists. Secondly, interviews were conducted with cyclists to understand their thoughts and real needs in respect of "environmental perception" and other aspects. According to crowd segmentation, this paper compares the behaviors of cyclists and non-cyclists, and analyzes different demands of cyclists. From the perspective of behavior research, this paper deepens the results of POE survey and finds out the cause of current issues.   A total of 300 questionnaires was sent out, and 282 of which were successfully recovered. There were 272 valid questionnaires, with an effective rate of 90.7%. Among the valid questionnaires, 58.4% (158) were "cyclists" in the experimental group and 41.6% (113) were "non-cyclists" in the control group. Sample composition is shown in Table 5. (1) As shown in Figures 6-9, Tangdaowan Park plays an important role in promoting residents to participate in outdoor activities, especially for residents who live within 10 kilometers radius. The cyclists in the park are mainly students, adults over 30 years old and children. Most of them choose spring and autumn to carry out   garden activities, which are greatly affected by the seasons. Most cyclists ride more than 2 or 3 times a month. Figure 10. Like non-cyclists, more than 60% of them ride in the   "South and North Straits" park. The distribution of different park visitors in the inland areas is shown in Figure 11. The demand of cyclists is different, leisure cycling activities and commuting cycling activities have higher requirements for entertainment, and the cycling environment of Tangdaowan Park can meet most of their basic needs. Sports cycling has a high demand for safety, and the current condition does not satisfy the user's requirements.

The distribution of different park visitors on the north and south shores is shown in
3. The daytime distribution of visitors is shown in Figure 12. The results show that sports cyclists are concentrated in the morning and evening, recreational bicycles are distributed throughout the day, and traffic cyclists are concentrated in the evening. Cyclists and non-cyclists show a similar pattern of time in visiting the park, and they have higher safety requirements of the cycling environment at night. The difference between the cyclists and non-cyclists is that the cyclists exhibit a relatively distributed pattern of visiting time, while the non-cyclists mostly choose to visit the park in the afternoon. The visiting time of different types of cyclists was observed in Tangdaowan Park (Figure 13), and the result shows that most sports cyclists visited the park during the morning and evening, whereas most casual and commuting cyclists visited the park after 5:30 p.m.
Generally speaking, the cycling experience of Tangdaowan Park is very satisfactory, especially under the background of the relatively low level of urban cycling environment construction. However, with the rise of "green lane and cycling fever", the cycling environment of the park must be further improved to better meet the increasing demand of citizens.
The social influence of cycling should be improved through the design of cycling environments to promote cycling as a lifestyle. The advantages and disadvantages of cycling environment can further strengthen or weaken the above functions. Further research will be done in the future, according to the spatial requirements, key points of use, areas to be strengthened and relevant design methods of cyclists.

Discussion and results
This section is based on the deficiencies found in the analysis and evaluation (POE and AHP) in the second section (Methodology) and the behavioral rules found in the questionnaire in the third section (Research Object Selection), summarizes the principles that need to be followed in the design of the urban park cycling environment. Based those principles, the various components of the cycling environment in the park, such as the cycling road, landscape environment, service facilities, and night cycling environment are analyzed and optimized. The space that is needed by different types of cyclists and non-cyclists are responded with proper solutions. In short, through the previous research and analysis, we understand the pros and cons of Tangdaowan Park in various aspects. This section will try to analyze the elements involved in "cycling environment design", and then propose corresponding strategies. Finally, based on the design plan, this section will also suggest some design principles to optimize the cycling environment of Tangdaowan Park.

Design contents of park cycling environment
Aiming at the specific case of "cycling environment in Tangdaowan Park", this paper studies the design of urban cycling environment from the aspects of design elements, design principles, design content and specific optimization methods, and then gives specific optimization results.

Design elements of park cycling environment
The design of cycling environments needs to consider natural and social elements. Among them, natural elements include topography, geomorphology, hydrology, climate, which is the basis of cycling environment. Social elements, including social attitude, social identity, cycling development law, are the soft elements of cycling environment.

Design principles of park cycling environment
In order to properly deal with the above elements, the design of cycling environments needs to follow four principles: "Systematic principle", "People-oriented principle", "Refinement principle", "Ecological economic principle". Firstly, the systematic principle requires that the design of cycling environment needs to meet the overall needs of the city, to meet the possibility of future development of cycling activities, and to improve the integration of environmental content; Secondly, the people-oriented principle requires the cycling environment to be user-oriented, and its characteristics and needs should be the first for the main population in cycling group; Thirdly, the refinement principle requires the cycling environment design to subdivide the cycling population and provide different environments according to different cycling needs. Finally, the ecological economic principle requires the cycling environment to be in line with the site conditions and pursue the comprehensive effects of ecology, society and economy. In the construction and subsequent optimization of Tangdaowan Park, the stacking of natural banks, the application of overhead platforms, the laying of gravel pavements, and the planning of high green areas are all based on ecological protection (Figure 14).

Design optimization of park cycling environment
The evaluation system of the second section divides the Park Cycling Environment into four aspects: Cycling Road System, Landscape Environment System, Service Facility System, and Night Cycling Environment, and then subdivided into 19 detailed indicators. Afterwards, the evaluation results pointed out two deficiencies of "Safety and Convenience" and three indicators (route selection and space-form construction, service facilities and information marking system design optimization, night cycling safety construction), which requires optimization. This section will improve the design of the Park Cycling Environment based on the above problems.

Cycling road system
Based on the findings of the second and third sections, the design of the Cycling Road System should pay attention to the safety of the road and the interest of the space, and separate different users through physical measures to improve the compatibility of the road space.
(1) Optimization of Line Layout As shown in the third section, the negative feedback on the cycling route mainly comes from the group of sports cyclists, that is, "After many rides, the cycling route is slightly monotonous, and it is better if more routes are selected." Through our observation, sport riders are much more frequent users comparing to leisure riders, and they are more sensitive to this problem.
The narrowest place on the south bank of the park is about 160 meters, and the widest place is over 1200 meters. The narrowest place on the north bank is over 120 meters, and the widest place is over 360 meters, which is wide enough to set more than two parallel cycling routes instead of only one. At the junction of the park on the north and south bank (Figure 15. Area B), the area and depth are large, but the design of cycling road is not considered. According to our observation, although the north park is narrow in width, many cyclists choose to ride on the sidewalk of the side of the north shore highway (Figure 15. Area A). Therefore, in order to optimize the route to form a satellite layout, circular routes can be set up at along the north shore park, and at the junction of north and south shore to optimize the satellite layout.
Secondly, with the help of the landscape axis of the south bank and the leisure area of the north bank, the connection between each branch should be increased to form a multi-ring layout. The final optimization results are shown in the Figure 16.
2.The Optimization of the Spatial Form of the Slowline in the Coastal Area The behavior patterns found in the third section: Sports cycling activities have high requirements for safety, and there is a big gap in demand. However, the cycling environment in Tangdaowan Park cannot meet this requirement in this respect. The conflict between Figure 15. Layout diagram of cycling line before revamping.
the activity time and space of cyclists and noncyclists is the main reason for reducing the quality of cycling environment. The profile of the current coastal walkway and the space on both sides is shown in Figure 17, and it indicates that there is sufficient space for renovation along the seashore. Therefore, in the design of the coastal walkway, we can add a new platform as the coastal bike line at  a lower level of the existing pathway, which is paved with granite for walking. Considering the increasing number of riders, the width of the bike lane should be more than 4.2 meters and the height should be at least 2 meters. The walking pathway and new bike lane are partially integrated at some locations, and physical measures are adopted to force cyclists to slow down at intersections. Viewing platforms can be extended from the walking pathway into the sea for viewing, fishing, playing chess and other activities. Cycling service facilities can also be set at the lower part of the platform to increase the fun of cycling experience.
As the park is built around the bay, hydrological conditions are equally important. The average sea level of Tangdaowan Park is −0.19 meters, the average tidal level is 1.80 meters, and the highest tidal level can reach 2.51 meters. In order to ensure that the park environment is not affected by the waves, the planned elevation of the coastal walkway is between 3.8 and 4.1 meters. The minimum viewing platform is set to 2.51 meters to ensure the safety of the park environment.

Landscape environment system
The Landscape plantation system should be dynamic and the various landscapes should be emphasized by using different color, line, scale, shape, texture and some other aspects. Water landscape should strengthen its relationship with roads and hydrophilicity.

The optimization of the landscape of the water in the park
According to the interview results in the third section, the major entertainment experience for sports cyclists in Tangdaowan park is scenic viewing. In contrast, leisure cyclists are more eager to stay on the hydrophilic platform, and their response is the strongest. Therefore, the lack of hydrophilicity of the landscape around the water body is the main problem at present.
The solution of this problem is to optimize the relationship between roads and the sea. The concave shoreline is easy to form circular routes and increase the sense of place. Streams and roads should be arranged in parallel to enrich the scenery on both sides of the road and this will provide maximum scenery experience. The bike line of the coastal walkway can also partly cross the water or river (Figure 18).

Service facility system
According to the interview results in the third section, many cyclists said: "It is very interesting to see other activities (such as square dancing, singing competitions, etc.) during cycling." However, open spaces in the park are usually separated from the cycling paths, and the surrounding areas are not suitable for parking bicycles.
The solution is to rearrange the spatial relationship between the open space and the bike lane system. The major principle for the change is to create a strong visual connection between the two by redesigning the height difference, terrain, vegetation and so on. This  can both increase the attractiveness of cycling activities to ordinary citizens, and also improve the cycling experience. At the same time, parking facilities and spaces should be set up in each square and nodal space. This not only meets the requirements of cyclists for the use of entertainment facilities, but also meets the needs of users of entertainment facilities with cycling behavior. Status quo and the final optimization results are shown in the Figure 19.

Night cycling environment
According to the interview results in the third section, it is found that the cyclists are concerned about the safety of the night cycling environment, and the lighting problem is the most prominent. The night cycling environment can be largely improved by integrating lighting design with other systems to ensure the use of the park at night.
First of all, the yellow LED light belt outside the coastal walk should be moved from the side wall of the driveway to the top of the boundary line, which meets the needs of urban landscape and the function of boundary calibration. Moreover, the park is usually affected by heavy fog as it is close to the sea, and thus the width of light belts should be widened moderately. The bright streetlights on the inner side of the coastal walkway should be moved to the outside, while the inner side should be provided by dim lighting, it will guide cyclists to avoid danger.
Secondly, in addition to the high street lights, there are a lot of low lighting. Because these lighting is close to the ground, it has a better lighting effect on the walking experience, so it should be widely used in the design of cycling environment. At present, the color of the low lighting is mostly white and yellow, and they can create uncomfortable and dazzling effects. There are also good design cases in the park to solve this problem, such as low lighting through plants, bushes, reflective boards to form a diffused reflection to better avoid the glare phenomenon. However, it can only be found in a few places, so it is necessary to apply this method in other places in the park. In addition, providing soft background lighting through the high level light source is also one of the solutions, but considering energy saving, it is not suitable for large scale application.
Thirdly, there are many objects that can isolate spaces in the park, such as steps, short columns, railings and other facilities. In the design of these components, the integrated design method is adopted, and the illumination is accessed to ensure the night marking isolation effect. Signs should be seen at night through the unified design of signs and lighting.
Finally, a local lighting design is needed for this kind of site. More importantly, it is possible to improve the lighting efficiency by combining local lighting design with some public spaces.

Conclusions
Based on "post-occupancy evaluation", this paper innovatively puts forward a new method of cycling environment design through three steps: "postoccupancy evaluation -behavior analysis -environment optimization". By using this method, this paper investigates, analyzes and optimizes the cycling environment of "Tangdaowan Park" in central Qingdao, and discusses the selection of specific methods (such as AHP, scale method, interview method) with practical operation, and verifies the scientific, systematic and operable of this method. At present, cycling activities develop rapidly and tend to be diversified. This new method can be widely used in the relevant projects. It can provide a good reference for the construction of cycling environments from both theoretical and practical perspectives, and meet the demand of cycling under the background of the new era, indicating a wide range of social significance in practice.
The results of this study showed that it is important to determine the cyclists perceptions and preferences that are related to their use of urban Greenways. Aiming at the specific case of "cycling environment in Tangdaowan Park", this paper studies the design of urban cycling environment from the aspects of design elements, design principles, design content and specific optimization methods, and then gives specific optimization results in "Cycling Road System, Landscape Environment System, Service Facility System, Night Cycling Environment". The results (such as the composition and weight of the cycling environment in Table 2) can directly provide support for relevant theoretical research and engineering projects, and it can also provide references for general application of this method.
In addition, this paper has obtained firsthand research data, which records and summarizes the information about "cycling group, cycling behavior, and cycling environment". The results are universal and representative to a certain extent. Cyclists among urban residents are mainly men, mostly middle-aged people and teenagers, and their residences are mostly less than 10 kilometers away from the cycling site. Cycling activities are more regular, and the intensity is higher than general outdoor activities. They are more affected by the season, and there are more cycling activities at night. In the design of the cycling environment in the park, it is necessary to focus on the psychological characteristics and needs of men and young people. The shading devices and night lighting of the cycling environment also need to be addressed.
The lack of "Safety and convenience" is the most prominent problem in the green park cycling environment at present. The problem of the spatial design involved in these issues is mainly concentrated in the following three aspects: Cycling Road System, Service Facility System and Night Cycling Environment. There are three suggested improvements: route selection and space-form construction, service facilities and information marking system design optimization, night cycling safety construction.
The conflict between the activity time and space for cyclists and non-cyclists are the main reason for reducing the quality of cycling environment. The construction quality of cycling environment should be improved to better serve all urban residents. If the conditions permit, the cycling environment should be taken as the independent design object, and the space experience of the cyclist should be taken as the core, and the road, landscape, greening, lighting, service station, infrastructure, information signs, traffic transfer and other relevant design contents should be integrated and integrated.