Abstract
“Green Transportation”, many initiatives as such promoting sustainable transportation choices have been adopted in major cities around the world. Public Bicycle including bike rental services have been considered on various occasions to solve the last mile problem and the consistently rising pollution levels due to increasing automobiles on the streets. However, bicycle parking stations have attracted very few researchers attention. Through this project I aim at understanding the reasons associated with bike parking facilities which has been perceived as the first and last mile resting points for the users .Higher occupancies translate as one of the contributors in the lifecycle of Bicycle as a mode of transportation whether public or rental services. The bicycle parking data for the month of “Nov 18 from stations located in Dublin city reported by the city council is used for carrying out the analysis. This provides the association of occupancy with the various aspects which might be critical to understand the users behavior towards the city council’s initiatives to promote a sustainable transportation. Analyzing this yielded an output according to which lighting , ease of access and medium security levels contributed largely towards higher occupancies .
INTRODUCTION
While bike sharing services have been around for more than a decade, the public infrastructure to promote cycling as a sustainable option to the conventional modes of transport has picked up lot of traction in the governmental policies . Dublin has recently had many such ventures with Dublin bikes being the most popular with the highest subscribers [2] . We consider the popular Bike renting model as a benchmark on whose data extensive research has been carried out. A typical Bike sharing system can be categorized in the following areas [3] (i) demand analysis (ii)Service analysis (iii)rebalancing operations. The analysis in this project aims at addressing the service analysis area through the available park stations data.
Little has been done on addressing the stations as these play an adequately important role along with other factors in the success of this Public mode of transportation, an initiative by Dublin city council. Firstly, we will check how the locations with different types of stations namely Sheffield, Railing and Hoops are performing in the existing areas and then the analysis of various factors affecting the occupancies of these stations. To validate the fields for the carrying out the statistical tests, Goodman Kruskal Gamma test for checking the co-relations and then proceeding for Anova to determine the factors which are strongly affecting the occupancy. This analysis helps in rebalancing the resources appropriately and have a better understanding of the user behaviour translating into better utilization and service.
Dataset
The dataset is from smartdublin.ie with recent updates from Nov 2018 which as noted earlier is by Dublin city Council. This comprises of cycle parking stations from across Dublin, with 556 records with exact locations. The Table 1. represents the attributes of the dataset.
EXPLORATORY ANALYSIS
Below image shows the summary of the first set of features being analysed for association with info on no. of cases processed. The Goodman Kruskal Gamma test is conducted as the variables considered are ordinal in nature.
Overall less than 50%of bike stands are situated near the entertainment centres. From the symmetric measure table, we see that the Gamma value is 427 which depicts a strong positive association with the type of stands, being statistically significant.
We see that around 60% (8) of railing stands are located nearby entertainment centres. From the symmetric measurements we observe that there is a strong negative association with type of stands and is statistically significant. The value of 1 is if the station is in the vicinity of a recreational establishment and value 0 is if it’s not in the vicinity of any recreational establishment.
Similarly, from the gamma test with security safety rating and type of stands, wherein security safety rating is categorised numerically as 0 – none, 1 – medium and 2 – high. Around 80% of the railing stands are rated as having no security, contrastingly 10% have been rated as high. For the case of Sheffield stands which constitute a major part of the total count have around 40% (189 stands) rated as no security and around 50% falling in the range of high security. From the Gamma value which is 0.305 we infer there is a positive strong association with type of stands and statistically significant.
We infer that 90% of the railing stands have in high lighting range and 10% in the no lighting range. The Sheffield stands have around 60% in high lighting range and around 39% in the no lighting range. The stainless-steel curved stands are having around 98% in the high lighting range. With the gamma value of around -0.312 we infer that there is a negative association with the type of stands and is statistically significant. The values are assigned as follows: 0 – none , 1 – low , 2 – medium , 3 – high .
We infer that railing stands are all in high condition, which can be categorised as in good build condition. The Sheffield stands having around 69% in high condition and around 31% in medium condition. All the stainless-steel curved stands are in high condition. From the gamma test value -0.437 we infer that there is a strong negative association with the type of stands and statistically significant. The values of physical condition are assigned as follows: 1 – medium, 2 – high .
Save your time!
We can take care of your essay
- Proper editing and formatting
- Free revision, title page, and bibliography
- Flexible prices and money-back guarantee
We observe that 80%(40 stands) of the railing stand stations aren’t located anywhere near a restaurant establishment. Sheffield stand stations having around 72% not anywhere close by to restaurant establishment. A whopping 72% aren’t anywhere near a restaurant establishment. The gamma test value computed is -0.36 which indicates weak negative association and the significance at 0.772 we infer that its statistically insignificant.
ANALYSIS OF FACTORS AFFECTING OCCUPANCY
Based on the exploratory analysis done the further analysis at this point concentrates on the strong factors affecting the occupancy of the stands. A pattern recognized for this would greatly help the council to concentrate on how the stands are being perceived. The addressing of the user behavior towards the stands will contribute largely towards the success of the initiative. This would also contribute in efficient decision making of the investments going into the stand and how well is it being received by the public.
A. Techniques : Anova
Anova accordingly would be an appropriate technique over the others as the dependant(outcome) variable is continuous and the independent(predictor) variables are ordinal. We will be carrying out statistical tests to support our findings from the exploratory analysis. We conducted Anova tests with the strong and weak associates to support the analysis and find their significant effect. We compare the means of more than one groups / categories. We will try to find the percentage of effect on the occupancy of stations.
B. Mean of occupancy Vs Ease of access
Anova analysis shows that occupancies of bikes are different among the ease of access categories ranging from Poor to Good. However, Poor and fair ease of access conditions show a significant relationship and might belong to same group as per the ANOVA. The mean graphs clearly depict that most of the occupancies have happened in good ease of access category rather than in poor or fair access range. So based on the overall stats we can support that good ease of access has primarily been one of the driving factors for rise in occupancy.
C. Mean occupancy Vs lighting
From the below table we can say with 99% confidence that there is significant effect on the lighting of the stands contributing towards higher occupancy.
Dissecting further the ANOVA and post hoc Tukey Honestly significant difference tests we can generalize that the means of the occupancies of bikes are different among the groups of the lighting range from none to high. According to Tukey HSD there is only significant difference between the groups of high and none lighting range. However, none and low lighting conditions show a significant relationship and might belong to same group as per the ANOVA .i.e. the occupancies in such lighting conditions is seen to be similar which fall in low category of occupation. The mean graphs clearly depict the most of the occupancies happened in medium and high conditions rather than in none and low.
D. Mean of occupance Vs Security safety rating
From the above graph of mean occupancy measure Vs security safety rating it can be inferred that there is a considerable occupancy rise when the rating is medium as when compared to the none rating. While there is no much significant rise of occupancy when the rating goes from medium to high.
RESEARCH AND INVESTIGATION
Bicycle as a mode of transport has always been perceived as the solution for the last mile travel in commuting, leisure etc. The city corporations and corporate industries as such have been continuously promoting these initiatives for healthier and cleaner environment. A major step towards the promotion of public usage by the inhabitants is through a strong infrastructure which would be the backbone supporting these initiatives. Parking stations have their own contribution in this area with aspects such as proximity between them , safety and many such factors. [4] A recent report by Dublin city council touched upon the councils initiatives to promote a sustainable public transport , recent installation of new parking stations and conversions of existing car parking spaces into bike parking stations to promote and accommodate more bikes. [5] The reported highlighted that adequate cycle parking stations plays a key role in the success of this scheme. [5]
Lack of identity of the bike parking stations leads to loss of people for this mode of commute [4]. An investigation by R.Buehler found that work places with better cycle parking facilities saw a 9% increase in bike commuting and bike parking had a strong statistical significance to bike parking. This paper employs multiple regression to examine the relationship of bike commuting, employer incentives and trip end facilities(bike parking , cyclist showers) which found only 1.7% of bike commuters . Employees with bike parking facilities at the trip end were associated with 1.78-time greater likelihood to cycle to work than those without any. Overall in the research area the level of evidence on the role of bicycle parking is limited, the parking facilities, conditions and qualities appear to be associated as determinants of bicycle commute for current and potential cyclists. [6] A research on the bicycle parking facilities as a contributor to the bicycle infrastructure driving the cycling initiatives found that high quality and frequent bicycle parking play a role in encouraging the public to switch to this sustainable mode of transport. [7] This paper categorised the design factors to seven categories to benchmark the selected cities. Each individual category was broken down to 50 criteria’s which were scored based on a scale of 1-5 by two riders. The research on examining the facilities at bike park stations affecting the bike commuting is seldom considered and very less in this area has been done
CONCLUSION
This project aims at finding the contribution of various factors associated with a bicycle park station. At stage(i)we conduct Goodman Kruskal gamma to measure the strength and association of the factors with the type of stands. This gives a considerable insight into the descriptive statistics part. At the stage (ii) Anova is conducted with strong associated groups to determine the individual subgroup effects. The results show that there is a considerably variation in the occupancy when the lighting is high and ease of access is good as compared to thecir the other scales of the respective factor. The security which is majorly perceived as a strong aspect has considerable effect when the rating of the stations is around medium and we see a nominal rise when its rated as high. The consequences of bike parking facilities are usually underestimated but the case here wherein occupancy which to an extent translates to this sustainable mode of commute being recognised by the citizens are highlighted. Thus, the tangible aspects pertaining to the bicycle parking stations does play a key role in the behaviour of the users.
Stuck on your essay?
