A study of influence zone for railway stations of different hierarchy in the Indian Railways: analyzing passenger travel patterns

ABSTRACT

The transportation planning of Indian Railways has become increasingly complex due to the growing population and increasing demand for mobility. To develop an efficient and effective last-mile transportation system, it is important to identify the extent of the influence zone of railway stations. This can guide policy decisions and infrastructure development. This research paper focuses on mapping the influence zones of 40 railway stations across the Northern Railway Division of India. It uses the travel length frequency distribution to analyze the distribution of passengers’ travel distances. The study shows that influence zones for railway stations vary significantly based on their hierarchy level. Higher-hierarchy stations tend to have larger and more dispersed influence zones, while those of lower-hierarchy stations tend to be smaller and more localized. The findings can be used to guide the expansion of existing stations and the improvement of transportation services. Overall, this study provides a valuable contribution to the field of transportation planning for Indian Railways. It demonstrates the importance of influence zone mapping in developing an efficient and effective transportation system.

KEYWORDS:

• Railway station influence zone
• univariate clustering
• last mile connectivity
• accessibility

Introduction

The influence zones of railway stations are geographic areas that surround the stations and are directly impacted by the station and its related transportation facilities. These areas are distinguished by their high level of accessibility to the station and can significantly affect the development of surrounding regions. The concept of influence zones is a crucial component of transportation planning as it assists in identifying areas that are most likely to be impacted by the development of railway infrastructure and services. Understanding the influence zones of railway stations can aid in the identification of transportation and urban planning needs, facilitate efficient resource allocation, and promote sustainable development.

The need to perform this study is to identify the influence zones of different hierarchies of stations, which eventually help in formulating policy for last mile connectivity and improve accessibility in the influence/catchment zone, this type of study was never conducted in Indian geographic conditions and on Indian Railways.

Numerous studies have been conducted worldwide to identify the characteristics and dimensions of influence zones for various types of railway stations. In a study conducted by Li et al, the influence zones of high-speed railway stations in China were identified by using a combination of GIS techniques and travel demand modeling (Li et al., 2019). The study found that these influence zones are characterized by a high level of accessibility to the station. A study by Silva et.al signifies that Improved rail service enhances accessibility, which in turn leads to positive effects on labor participation and migration (Silva et.al, 2013). Improvement in accessibility reduces the time space frictions, which increases the interactions between eco-nomic activities, and this depends on the influence zone of the railway station. The rail mode has a more dominant role in public transit as compared to the bus mode. Furthermore, zone-based travel patterns were studied, which impact public transport and accessibility to the stations (Choi et al., 2020). In summary, the identification of influence zones around railway stations is a critical aspect of transportation planning. Influence zones provide a means of understanding the spatial distribution of demand for railway transportation, and they can be used to guide the development of infrastructure and services. The characteristics of the influence zones vary depending on the location and type of railway station, as well as the surrounding areas’ features. The following sections of this paper will present a detailed analysis of the influence zones of railway stations in a specific geographic area. This analysis will focus on identifying the characteristics and dimensions of influence zones for different types of railway stations using travel length frequency distribution.

Understanding the concept of influence zone

The majority of the work employs the Euclidean method, which appears to be the ‘golden rule’ (Connor & Harrison, 2012), and is primarily based on walking access to the station. However, there is a lack of literature on the territorial influence of a railway station, which should also be defined by its driving accessibility).

A train station’s territorial impact is measured in various geographic regions (Connor & Harrison, 2012). There are several references to attractiveness areas, and influence areas. These terms all refer to the same thing: the region surrounding a station. Yet their meanings are not the same. In several studies, rail operators therefore use the phrase ‘influence zone’ to describe the region where individuals are likely to use the train. Their reasoning is predicated on estimating probable ridership

Researchers like Cervero use the term ‘attractiveness area’ to describe the region where residents may benefit from boarding the train (Cervero, 1995). In this assessment, Cervero identified the perimeter of the station catchment based on the acceptable amount of time for passengers to access the railway station. In the case of Bahn Ville, two approaches for influence zones were established. The first was the buffer zone, a disc centred on the station with a fixed radius. The second was the isochrone, a network-based boundary that takes mobility restrictions into account; both of these approaches were also followed in this study (Hostis et al., 2010). A train station’s territorial influence extends beyond this proximity zone. It should also be defined by its accessibility by vehicle. Shen, et al. suggested that for each railway station area (or administrative region), the impact of the station should be studied both internally and externally. The impacts of railway stations are quite diverse even with similar city sizes on the same network (. Influence zones can be defined on the basis of distance and travel time. Some of the research is in Table 1 Quantitative Assessment of Zones for Railway Stations (Shen et.al 2017).

Table 1. Quantitative assessment of influence zones for railway stations.

Factors for Influence zone	Quantitative assessment of influence zones for railway stations	Reference
Distance Based	3 km Radius - France	French Transport Firms (Govt. of France)
	5 km Radius – France	(Cervero, 1995)
	Distance of 7 km for stations – Canada	(Ermacora et al., 2013)
	80% of railway stations have an influence within 6 km in Europe	(Hasiak & Richer, 2012)
	6 km radius to 8.5 km radius	(Hasiak & Bodard, 2016)
	For Regional Rail Transit system o Core Station Area- 400 m o Primary Influence zone – 800 m o Secondary Influence zone – 8000 m	(Neff & Dickens, 2009)
	800 m Influence zone	(Ohland & Dittmar, 2004; Untermann, 1984)
	4500 m and 8700 m in Philadelphia and San Francisco, respectively	(Bradley J & Rivasplata, 2014)
	1.96 and 2.13 km for home-to-station and station-to-work trips for, respectively	(Zhao & Li, 2017)
Travel Time based	Driving accessibility in the range from 15 to 20 minutes around the station	(Hostis et al., 2010)
	40 min driving accessibility	(Hasiak & Bodard, 2016)
	5-min walk	(Bernick & Cervero, 1997)
	10-min Walk	(Calthorpe, 1993)

Methodology

For this study, pilot cases were selected in the Northern Railway division, where primary surveys were conducted, which included the origin and destination surveys of the boarding and alighting passengers and further built the conclusion on the extent of the influence zone. For this study, the categorization of Indian railway stations was selected based on their importance in the region: number of platforms, number of passengers, number of train stoppages, type of trains, and area of the station, as suggested by (Bhatnagar & Ram, 2023). The impact was then studied at all the stations selected in the sampling study. The stations primarily selected for the pilot study are Sonipat, Karnal, and Kurukshetra under Category 4; Panipat and Amritsar under Category 3; Ludhiana, Jalandhar, and Ambala under Category 2; and New Delhi under Category 1. All are part of the Northern Railway Division of India. The scope of the influence zone was decided based on the travel-length frequency distribution.

Origin and destination surveys were conducted at all the sample stations. A priori method is used with an effect size of 0.35 and an error probability of 5%. The total number of samples that need to be collected is 209 at each selected station (Figure 1).

Figure 1. Sample size determination for Passenger OD surveys.

Identifying the extent of the influence zone for different hierarchy of stations

From the study (Bhatnagar & Ram, 2023), it was evident that stations needed to be categorized based on their importance in the region. The influence zones are divided into three sections based on (Neff & Dickens, 2009).

• Core Zone: Based on the Lower Quartile (25^th Percentile)

• Primary Zone: Based on the Median Quartile (50^th Percentile)

• Secondary Zone: Based on the Upper Quartile (75^th Percentile)

During the primary study, passenger characteristics such as the purpose of the trip, the amount of luggage carried, and the mode of access were observed.

The importance of the station in the region and on the rail network has a huge impact on the availability of access modes, which eventually determine the extent of the influence zone. When using TLFD to compare stations in categories 4, 3, 2, and 1, it was discovered that stations in similar categories exhibit similar behaviours. It was observed that stations with lower importance (category 4) in the region tend to have a smaller influence zone, are usually accessed by walk, 2-wheeler, and IPT, and that 70% of the trips by passengers were work trips, unlike category 1 stations, which were distributed among work and education trips (Figure 2).

Figure 2. Comparison of passenger characteristics for Category 4 and Category 1 Stations.

Further category wise TLFDs were prepared to identify the extent of influence zones:

1. 4 stations (Lowest in the category list)

Category 4 railway stations are the lowest in the hierarchy of railway stations in India, and they showcase similar passenger behavior based on Figure 3. This is because these stations are in regions where there is a lack of last mile connectivity modes such as public transport systems, making it difficult for passengers to travel further from the station Table 2. As a result, most passengers at these stations are either locals who walk to their destinations or those who have pre-arranged transportation waiting for them. Moreover, these stations are in regions where the importance of rail travel is not as high compared to other regions, resulting in fewer passengers overall.

Figure 3. Trip Length Frequency distribution for selected Category 4 Stations.

Table 2. Influence zones based on primary study.

Zone	Sonipat (Catchment in Km)	Karnal (Catchment in Km)	Kurukshetra (Catchment in Km)
Core Zone (25 Percentile)	1.6	0.8	1.8
Primary Zone (50 Percentile)	2.1	2.3	2.5
Secondary Zone (75 Percentile)	3.1	4	4.2

2. Category 2 and 3 Stations

On comparing categories 2 and 3 (Figure 4) (Figure 5), the average travel length of passengers from category 2 stations was observed to be an average of 3.3 km more than that of category 3 stations (Table 3 and 4).

Figure 4. Trip Length Frequency distribution for selected 2 and 3 Stations.

Figure 5. Trip Length Frequency distribution for selected Category 2 station - Ambala.

Table 3. Observed Influence zone for Category 3 Stations.

Zone	Panipat (Catchment in Km)	Amritsar (Catchment in Km)
Core Zone (25 Percentile)	2.4	1.7
Primary Zone (50 Percentile)	3.2	4
Secondary Zone (75 Percentile)	3.4	4.3

Table 4. Observed Influence Zone for Category 2 Stations.

Zone	Ludhiana (Catchment in km)	Jalandhar (catchement in km)	Ambala (Catchment in km)
Core Zone (25 Percentile)	3.5	3.8	6.2
Primamry zone (50 Percentile)	6	5.7	7.4
Secondary Zone (75 Percentile)	7.1	6.8	7

3. Category 1 Station – New Delhi Railway Station

In India, Category 1 stations are the highest category stations in terms of regional importance; these stations are the largest and provide regional connectivity. After the primary survey, it was observed that the influence zone threshold is 18.5 km (Figure 6). While the median is 15.5 km and the core zone influence is observed to be 5.3 km, the major reason is the availability of the metro system, which enhanced the reachability and accessibility of the access passengers.

Figure 6. Trip Length Frequency distribution for selected Category 1 station – New Delhi.

Sample size determination

A sampling study was done to establish the number of stations for which the study was to be conducted. A-priori power analysis was used to calculate the sample size given some estimated effect sizes, alpha, and power. Using an a priori power analysis, a sample size of 40 (Figure 7) with 6 predictor variables (Bhatnagar & Ram, 2023) will provide us with an R-square value of 0.95 and an error probability of 5%.

Figure 7. Sample size determination for Station selection.

Result

This study was conducted on all the 40 stations and influence zones were determined. Further, these influence zones help in the identification of the accessibility to stations of different hierarchies (Table 5).

Table 5. Influence zone assessment for different Hierarchy of stations.

Stations	Railway Station Category	Importance of the station (Bhatnagar & Ram, 2023)	Contour Catchment	Average Secondary Influence zone (Km)
New Delhi	1	0.83	18.5	16.9
Anand_Vihar	1	0.54	18.1
Old Delhi	1	0.73	17.8
Nizamuddin	1	0.54	16.8
Lucknow	1	0.55	13.2
Sarai_Rohilla	2	0.43	11.2	8.4
Ghaziabad	2	0.45	8.5
Ambala	2	0.53	7.7
Varanasi	2	0.46	7.6
Ludhiana	2	0.44	7.1
Jalandhar	3	0.36	6.8	3.90
Amritsar	3	0.4	4.3
Jammu_Tawi	3	0.35	4.25
Bathinda	3	0.35	3.85
Chandigarh	3	0.36	3.65
Bareilly	3	0.34	3.65
Patiala	3	0.37	3.5
Panipat	3	0.34	3.35
Haridwar	3	0.36	3.35
Moradabad	3	0.42	3.25
Saharanpur	3	0.39	2.9
Faridabad	4	0.27	3.5	3.1
Rohtak	4	0.33	3.49
Gurgaon	4	0.29	3.4
Muzaffarnagar	4	0.29	3.35
Kurukshetra	4	0.31	3.2
Sonipat	4	0.3	3.2
Meerut_Cantt	4	0.22	3.2
Raebareli	4	0.29	3.2
Rampur	4	0.25	3.15
Dehradun	4	0.25	3.1
Hardoi	4	0.3	2.9
Karnal	4	0.29	2.85
Amethi	4	0.23	2.85
Hapur	4	0.3	2.85
Meerut	4	0.33	2.8
Roorkee	4	0.27	2.75

After the identification of the influence zone, both the importance of the station and the influence zone are compared using Pearson correlation (Table 7), which is a measure of linear correlation between two sets of data. Pearson’s correlation coefficient is the covariance of the two variables divided by the product of their standard deviations (Table 6)

Table 7. Pearson Correlation among Railway station Category, Importance of the station and Contour catchment.

Correlation matrix (Pearson):
Variables	Railway Station Category	Importance of the station	Contour Catchment
Railway Station Category	1	0.814	0.807
Importance of the station	0.814	1	0.785
Contour Catchment	0.807	0.785	1

Note: Values in bold are different from 0 with a significance level alpha=0.01.

Table 6. Summary Statistics for the selected stations.

Variable	Observations	Minimum	Maximum	Mean	Std. deviation
Railway Station Category	40.000	1.000	4.000	3.027	1.067
Importance of the station	40.000	0.220	0.830	0.380	0.131
Contour Catchment	40.000	2.750	18.500	5.923	4.843

There is a strong positive correlation observed between contour catchment and the importance of the station in the region (Bhatnagar & Ram, 2023); as the importance of the station increases in the region, the influence zone of the station also increases. Another analytical analysis was performed using univariate clustering, also known as one-dimensional clustering, which is a technique used to group similar objects based on a single variable or attribute. This type of clustering is useful when working with data that has only one measurable feature. One common univariate clustering method is k-means clustering. A summary is presented in (Table 8) for contour catchment.

Table 8. Summary Table for contour catchment.

Variable	Observations	Minimum	Maximum	Mean	Std. deviation
Contour Catchment	40	2.750	18.500	5.923	4.843

Using K-mean clustering, 4 clusters were identified (Table 9) according to the category of the stations. ANOVA was performed (Table 10) which is a statistical test used to compare the means of two or more groups to determine if there is a statistically significant difference between them. Here, the data is observed to be significant.

Table 9. Category-wise Contour Catchments (K-Mean Clustering).

Category of Stations	Contour Catchment in Km
Category 1	17.90
Category 2	12.20
Category 3	07.54
Category 4	03.35

Table 10. Anova Test for k-Mean Clustering.

	ANOVA
	Cluster		Error
	Mean Square	df	Mean Square	df	F	Sig.
Station Importance	.156	3	.005	33	34.081	.000
Contour Catchment	278.267	3	.286	33	972.700	.000

Further, all stations in India or around the world within the respective categories can be assessed for accessibility and setting up new railway infrastructure using the contour catchments in Table 8. The station influence zone increases as the category of the stations increases from category 4 to category 1 (the highest category) in the hierarchy; this can be observed in (Figure 8), (Figure 11), where the influence zone is one of the critical parameters for station accessibility (Table 11, Figure 10).

Figure 8. Railway Station Importance vs Contour Catchment for Indian Railway Stations.

Table 11. Summary table of regression analysis.

Regression Statistics
Multiple R	0.886
Adjusted R Square	0.779
Standard Error	2.275
Observations	40

A relationship of Contour catchment and Importance of the station is studied using linear regression (Figure 9), where goodness of fit (R²) is observed to be 0.7854. Developing an equation using importance of the station as dependent variable and Contour catchment as independent variable.

(1) $y=32.676x-6.4943$(1)

Figure 9. Importance of the station vs Contour catchment.

Where $y$ is the contour catchment/Influence zone of the station and $x$ is the importance of the station in the region.

Figure 10. Box-plot for Contour catchment and Railway Station Importance.

Figure 11. Indian Railways Influence zone mapping.

Conclusion

The impact of a railway station extends beyond its immediate vicinity, and its importance is directly linked to the extent of its influence zone. Transportation planners must consider influence zones when expanding railway services and infrastructure, as they provide valuable insights into the areas that are most likely to be affected. The contour catchment analysis based on station importance and hierarchy can help determine the extent of the influence zone. It is clear from the analysis that stations with a lack of access modes, particularly in lower categories, tend to have smaller influence zones. In summary, understanding the influence zone of a railway station is crucial for effective transportation planning and infrastructure development.

Through equation (1)(1) $y=32.676x-6.4943$(1) , influence zones for different hierarchical stations can be identified, and infrastructure such as transit systems and bus stops can be developed for better accessibility.

While previous studies have primarily focused on transit ridership estimation, walking accessibility improvement, and land use change effects within a defined influence radius around a transit station, there is a need to broaden our understanding of a station’s catchment area. Specifically, further investigation is required to analyze the secondary catchment area that can be reached by the transit mode.

Through this study, the catchment zone for the stations was observed to be 3.35 km for the lowest category and 17.90 km for the highest category, in comparison with previous studies conducted around the world secondary influence zone ranges from 3.00 km to 8.70 km. Thus, the influence zone for the station should be increased for higher category stations.

Author contributions

In the current situation, Indian railways is growing at a rapid speed and expected to handle 15 billion passengers by the end of 2030, but with this increase in patronage there will be an impact on the infrastructure capacity associated with the station, especially with the accessibility infrastructure. The need of an hour is to develop a robust supporting infrastructure for an efficient operation of a railway station, for this it is important to identify the extent to which a railway station influences the region. This article discusses an appropriate influence zone for railway stations of different hierarchies in India, which will help policymakers and planners to develop a suitable infrastructure for accessibility.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Due to the nature of the research, supporting data is not available.