An Assessment on the Locational Pattern of Petrol Filling Stations along Lasu- Isheri Road Corridor

This study aimed at assessing the location and spatial distribution of petrol filling stations along LASU/Isheri Road, Ojo, Lagos state. The objectives are to map out all the petrol filling stations in along Lasu/Isheri road; to examine the volume of traffic along the road corridor; to determine the contribution of petrol filling stations to the traffic volume on the road, and to ascertain the road traffic challenges that are caused by the petrol filling stations (PFS). Geographical Positioning System (GPS) was employed to collect primary data; also, questionnaires and traffic count sheets were employed. The study found that the PFS along the road corridor is clustered with a Z-score of -7.34 and NNI of 0.440285. Also, the maximum peak hour volume was estimated to be 4198.6 pcu/hr. The PFS along the corridor are seen to contribute significantly to the traffic volume on the corridor. Finally, the dominant traffic challenges along the corridor include traffic gridlock which sometimes results into road traffic crashes which are triggered by the concentration of PFS in the study area, the proximity of PFS to a road intersection, overflow of the queue into the roadway, and to a minimal extent parking of tankers along the roadway and lack of setback. This study suggests strategies that can be adopted for locating PFS to ensure the free flow of traffic along the road corridor where they are located.

. Ojo LGA, on the other hand, has an estimated population of 598,071 (NPC, 2006), and is located between latitude 60 22′N and 60 32′N and on longitude 30 4′E and 30 20′E. The local Government Area is bounded by six other local government areas and is among the seven Local Government Areas occupying the coastal plain of Lagos metropolis.
The predominant land uses in this area are commercial, educational, religious, recreational, and residential land-use. LASU/Isheri road is a four (4) lanes dual carriageway with a width of 3.65metere per lane and an approximate length of 16km.
The predominant land-use types along the road are commercial, educational, religious, recreational, and residential land-use. The primary data for this study was sourced for through physical observations, traffic surveys, administration of structured questionnaires, and interviews, this is in a bid to identify the filling stations within the study area. Data such as the coordinates of the filling station with which the exact location, distance, and the spatial distribution of the filling stations can be ascertained LASU/Isheri Road would be gotten through the use of a GPS device. The traffic volume data of filling stations and the road understudy was also collected. Secondary data was gotten from published journals, unpublished thesis, and relevant texts.
For primary data, the population of the study includes all the filling stations along the road under study, based on the reconnaissance survey carried out on the study area and a total of 47 filling stations were counted along the road. The number of population is the basis for the population of the study. Table 1. Names of petrol filling stations in the study area A stratified sampling technique was adopted in the study. During the reconnaissance survey carried out on the study area, a total of 47 filling stations were counted and coordinate reading was taken with a GPS device. The data was extracted from the GPS and imported into the Arc-GIS environment through which a Nearest Neighbour Analysis was carried out to get the spatial distribution of the filling station. The spatial map indicated that some of the filling stations exhibit high clustering, medium clustering, and low clustering. Thus, the population was divided into three (3) strata; high, medium, and low clustering, samples were then randomly drawn from this each stratum. The sample size for the study was derived based on the recommendation of the Institute of Transportation Engineers (ITE) for trip generation rate study which states that "if analyst intends to establish a Also for the questionnaire administration, samples were drawn from the three (3) regions identified above; questionnaires were administered to the commuters at bus stops close to sampled filling stations. Therefore, questionnaires were administered at fifteen (15) bus stops across the road. Purposive sampling technique was adopted by the researcher in administering the questionnaire to sample commuters that are not in haste. Twenty (20) questionnaires were administered in each bus stop, 10 questionnaires were administered during the AM and PM peak periods respectively across the fifteen (15) bus stops, to realize three hundred (300) commuters sampled. The sample frame and size are summarized in the table below.

Nearest Neighbour Analysis
This is the spatial analysis of the distance between a point and its closest neighboring point to determine if the point pattern is random, regular, or clustered. The nearest neighbor index is expressed as the ratio of the observed distance divided by the expected distance. The expected distance is the average distance between neighbors in a hypothetical random distribution. If the index is less than 1, the pattern exhibits clustering; if the index is greater than 1, the pattern is tending towards dispersion. The nearest neighbor index produces results that range from 0 to 2.15, a result of exactly 0 depicts clustering, 1 depicts randomness, and 2.15 depicts regularity.
The nearest neighbor index (NNI), Rn is given as:

= 2ᵭ√
Where: Where: Vehicle Trips = the total of all trips entering plus all trips exiting a site during a designated period.

Results and Discussion
Spatial Distribution/Pattern of the Filling Stations along Lasu/Isheri Road From the survey carried out, there are 47 functional filling stations along the road corridor of study which is 16kilometers long.
The survey was centered on the proximity of filling stations from one another along the road. ArcGIS 10.3 Spatial Statistics extensions were used to derive the distance between each feature centroid to that of the nearest feature (filling station). The result revealed that the observed mean distance between adjacent or close filling stations is 162.24 meters as opposed to the expected mean distance of 368.49 meters as proposed by the Arc-GIS software. The Arc-GIS software also revealed that the pattern of the petrol station along the road is clustered with a Z-score of -7.34. That is, there is a less than 1% probability that the clustering is a result of random chance. From Figure 3 a Z-score of less than -2.58 describes a clustered pattern. From Table   3 below, the nearest neighbor index is 0.440285. If the nearest neighbor index is less than one (1), the pattern exhibits clustering, and if the index is greater than one (1) the trend is the dispersion or competitive. Thus, the clustered pattern can be said to be severe in its clustering since its derivation from zero (0) is slight. The Spatial pattern map of filling station showing the severity of clustering at a different location the road is shown in Figure 4. Embedded in the cluster are different degree of clustering which includes, High clustering, medium clustering and Low clustering, each of these clusters have varying degrees of impact on the traffic flow on the roadway.
It is a widely known fact that the filling station is one of the major attractors of traffic amongst other land-use types. The clustering of filling stations along the corridor has a negative implication on the flow of traffic in that since the expected distance between each station has not been adhered to. The haphazard location violates the fundamental objective of planning which is providing the right site for the right use at the right time for the right purpose to achieve spatial functionality, efficiency, and aesthetics. During periods of fuel crises, queues of vehicles at few stations that had fuel often resulted in severe traffic hold-up on an adjoining road. This has negative implications for social and economic activities in the city.  The volume of traffic along the corridor was converted to Passenger Car Unit which is a vehicle unit used for expressing highway capacity. One car is considered as a single unit, cycle, the motorcycle is considered as half car unit. Bus, truck causes a lot of inconvenience because of its large size and is considered equivalent to 3.0 cars or 3.0 PCU.

Type of Vehicle PCU
Car, taxi, pick up 1.0 Cycle, motorcycle, Tri-cycle 0.5 Bus, truck, 3.0 Source: Wikipedia Figure 5 and Table 4 reveals the volume per hour variation among the days of the week. The maximum / highest hourly volume was estimated to be 3862.7 pcu/hr which was recorded on Monday. The lowest or minimum volume for the week was observed on Saturday (2210.1pcu/hr). This shows that the traffic volume along the road is usually high during weekdays and low on the weekend. The results of the traffic volume on the road depict a high level of land use activities going on in the area. The volume of traffic along the road which is considerably high with a peak hour volume of (4198.6pcu/hr), reveals the reason why there is always traffic congestion on the road during peak periods, knowing high traffic volume have negative implications on traffic flow. It also explains the reason why there is a high demand on the petrol filling stations along the corridor necessitating in-flow and out-flow of traffic at filling stations which at times due to the haphazard siting of the filling station impedes traffic flow along the corridor especially during the period of fuel scarcity when there is usually spillover of queues to the roadway thereby reducing the carriage capacity of the road and consequently causing traffic congestion.  To ascertain the contribution of filling stations to traffic volume along the road, a trip generation rate analysis was carried out The trip generation rate of petrol filling stations The site selected for the study is chosen based on the region of the road that exhibits high clustering, medium clustering, and low clustering. Five (5)   AM(Ante-Meridian) Peak Period: Morning peak period.

Characteristics of Highly Clustered Filling Station
This explains the reason why there is always traffic gridlock around this region especially during periods of fuel scarcity. The volume of traffic attracted to the filling stations is high (due to large GFA and FP), queues from this filling stations tend to spill over to the roadway, the cumulative effect of this clustered station is seen in the severity of the traffic gridlock in this region.

Characteristics of Medium Clustered Filling Stations
The cumulative effect of the vehicular trip generated by these filling stations on the roadway during periods of high demands is mildly felt because they are not too clustered. However, in some cases, the impact is highly felt because the filling stations within the region have limited capacity in terms of GFA and the number of FP and are not able to efficiently service the teeming vehicular traffic it attracts. (see Table 6).  (2017) Characteristics of Low Clustered Filling Stations As seen in Table 4.7, the average GFA is estimated to be 28968.2ft 2 while the average number of fuelling position is 9, and the distribution of filling stations within this region are seen to tend towards dispersion. This explains the reason this region is mildly affected by the overflow of queue unto the roadway during periods of high demand. On average, the filling stations are said to have fair enough capacity to service the number of vehicular traffic it attracts. The trip data for the AM peak period and the PM peak period of the days' data were collected are shown in Tables 8, 9, and 10 for filling stations that exhibit high clustering, medium clustering, and low distribution respectively. The trip rates shown are rates per number of fuelling positions (FP) and 1000 square feet of gross floor area (sq. ft. GFA). These rates are calculated using specifically the data collected during the same day.

AM/PM Peak Perio
The Vehicular trip generation characteristics of the clustered filling stations along the road under study are examined through the site characteristic survey. The preliminary estimates of vehicular trips generated by this type of land use are obtained and early peak (A.M.) and late peak (P.M.)Vehicular Peak hour rates needed for traffic impact assessments are established. The independent variables used for this study are selected due to their measurability from the study site. Tables 6 reveal the vehicular trip generation rate of each independent variables, 21 vehicular trips were generated per 1000ft 2 of gross floor area, 37 vehicular trips were generated per fuelling positions. Therefore, all things being equal, a filling station with 12 fuelling positions would generate (12x37) 444 vehicular trips during a typical AM peak period. Also, looking into the percentage of directional traffic in and out of the filling station, it is revealed that the percentage of vehicles entering the stations are higher than the ones exiting during peak periods, the situation is worst during PM peak period where 61.01% of the trip to the station is recorded entering the station while only 38.99% was recorded exiting the station within the same hour. This suggests that the service rate is low within the region which thus leads to long queues at the stations and the adjacent road.
Because the region where these rates are generated from is clustered (i.e. the observed distance between the filling stations is short), the cumulative effect of trips generated by each filling station hurt the flow of traffic within the region, especially during periods of high demand where the trip to filling stations rise by almost 20-50%. During this period, cases of spilled queues unto the roadway may arise from almost all the filling stations within the region thereby resulting in serious traffic gridlocks which consequently make traffic movement through the region stressful, and time and energy-consuming.  (2017).

AM/PM Peak Period Trip Data for medium clustered Filling Station
The trip generation characteristics of the regions that exhibit medium clustering of filling stations along the road under study are examined through the site characteristic survey. As observed in Table 7, the trip generation rates per 1000ft 2 GFA and FP are 21 and 57 vehicular trips during the AM peak period and 17 and 45 vehicular trips respectively during the PM peak period.
Considering the percentage of trips that enters and exits the station within the same hour, it is revealed that the number of vehicles entering the station is higher than the ones exiting, especially during PM peak period were 58.46% of trips at the stations entered and only 41.54% exited within the same hour which depicts that there are always queues in the stations within the region.
The high volumes of traffic that the filling stations within the region experienced during the peak hour can lead to spillback into adjacent roadways and thereby increase the potential for traffic congestion and collisions and this potential is high for this region because the filling stations are to a certain degree clustered.

AM/PM Peak Period Trip Data for low clustered Filling Stations
The trip generation characteristics of the regions that exhibit low clustering of filling stations along the road under study are examined through the site characteristic survey. From the collected data, it was determined that the average trip generation rate during a typical AM period was 11.74trips per 1,000 square feet GFA and the PM peak hour was 9.07 trips per 1,000 square feet GFA. Likewise, the filling stations were found out to generate 37.8 and 29 trips per fuelling positions during AM and PM peak periods respectively. Also, the percentage of trips entering the station is higher than the ones exiting, which shows there is a queue in the stations. Despite the high volume of trips generated by the stations in this region and the rate of the queue at the stations, the impact will be minimally felt by the roadway because the stations are spaced and not close to each other. See Table   4.10 for details. Summary of the percentage contribution of filling stations to the traffic volume along with the road understudy Tables 11a and 11b reveal the contribution of filling stations to the traffic volume along LASU/Isheri road during AM and PM peak periods respectively. It is shown that filling stations' contribution to traffic volume during the AM peak period is higher than its contribution during the PM peak period for the high, medium, and low clusters.
However, the contribution to traffic volume differs with varying degree of clustering of the stations. The high clustered stations are seen to contribute more to traffic volume along the road corridor with 21.2% contribution rate followed by the medium clustered stations and low clustered stations which contributes 21.2%, 15.6% respectively, as seen in (Table 11c).
This exposes that filling stations along the road corridor contributes significantly to the traffic volume on the road. An increased traffic volume brings about a corresponding increase in traffic problems experienced on the road if it is not properly managed.
The areas where highly clustered filling stations are located are the worst hit by the traffic flow problems experienced during the AM peak period.    (2017) The concentration of filling stations Regarding the concentration of petroleum filling station in the study area, 81.7% of the sampled populations are of the view that the petrol filling stations are too many and close to each other on the 16km length of the road, while only 16.7% indicated that the filling stations are not too concentrated. This is a pointer to the reason there is always congestion on the road during periods of fuel scarcity and high demand. This clustering has a combined effect on the roadway.

Filling Station as Part of the Causative Factor of Traffic Problems
From the study, 91.7% of the respondents perceived that the petrol filling stations across the road contribute significantly to the traffic problems experienced on the road.

Predominant traffic problem experienced on the road
From the study, 69.3% of the respondents agreed that the predominant traffic problem caused by filling stations along with the road traffic congestion, which in turn result in delay and waste of productive man-hour, excessive burning of fuel, air, and noise pollution to mention but a few. Other problems include Road traffic crashes (24%), Fire accident (2.7%) and traffic law abuse (1%) Reasons for the observed traffic problems As gathered from the respondents, the reasons for the traffic problems on the road are as follows: a. As seen in d. The clustering of filling stations also adds to the traffic challenges experienced on the road. Petrol Stations along the road were noticed to be too close to each other; as supported by 78.3% of the respondent, some were even developed side-byside-side thereby resulting in a wider negative impact on the immediate environment. e. Wrong parking by tankers about to off-load fuels on the roadway also disrupt the free flow of traffic along the road.
Though this is not a common sight on the road, however, on days when it does happen, it results in a serious traffic bottleneck.
f. Also, 86.7% of the respondent indicated that often, the turning movement of petrol tankers and other vehicles in and out of filling stations impedes traffic flow along the carriageway.   Finally, as gathered from the survey carried out on the study area, it was revealed that the traffic impact of filling stations is not to be undermined judging from the result of the traffic problem analysis. The dominant traffic problems as gathered from the survey include, traffic gridlock which sometimes results into road traffic crashes which are triggered by the concentration of filling station in a region, the proximity of filling station to a road intersection, overflow of the queue into the roadway, Turning movement of petrol tankers and other vehicles in and out of filling stations, parking of tankers along the roadway and lack of setback.

Conclusion and Recommendations
This study has observed that the siting of petrol stations on the study road are not evenly distributed as some pattern show a high clustering, medium clustering and low clustering as shown by the nearest neighbor index analysis, the filling stations spread across the road are too many, there is over 47 filling station on the 16km long road and some are still under construction.
Judging from the number of trips attracted to the area by the filling stations especially the highly clustered ones, it can be said that the clustering of stations within a region has a negative implication on the flow of traffic on the roadway. Certainly, this is as a result of the utter disregard for planning regulation on the minimum distance that must exist between filling stations set by the department of petroleum resources (DPR) practiced by oil marketer and it may perhaps be due to the overt increase in human population and vehicular ownership and other domestic/industrial equipment that consumes fuel. This incessant increase in demand for fuel has resulted in the establishment and spread of fuel filling stations in towns and cities today. Even though there is a need to make sure that fuel is readily available in every region of the nation as it is a major source of energy and therefore important for the economic development of a nation, its negative impacts from the processes of fuel retailing should not be overlooked.
It is therefore pertinent to note that, location of filling stations along a major road corridor should be strictly regulated and monitored, due regards must be given to planning criteria, safety, and environmental concerns, knowing that citing filling stations haphazardly have a negative implication on traffic flow especially during the period of high demand on fuel.

Recommendations
To alleviate the identified traffic impacts of filling stations on the roadway and also for future development, recommendations are made based on the findings from the study and views of respondents toward inhibiting such impact. Recommendations made from the findings. There should be: i Strict enforcement to reduce clustered petrol filling stations; and ii Adoption of seamless traffic management techniques such as directional signs, etc. particularly around petrol filling stations.
The proximity of filling stations to U-turns/ road intersection Lack of setback of filling stations from the roadway.
Overflow of the queue from Filling station into the roadway especially during fuel scarcity