36-Full Length Article-236-1-10-20180705
36-Full Length Article-236-1-10-20180705
36-Full Length Article-236-1-10-20180705
11 (2013) 99-112
1. Introduction
Speed and travel time are the most commonly used indicators of performance
for traffic facilities. Network travel time contours, speed trends, maps of link
speed, and various displays in transportation management centers emphasize
the importance of these two parameters to the public and to the profession.
When facilities are congested, the amount of delay and the extent of the
congestion are common supplemental indicators. The terms “delay” and
“congestion” are almost always synonymous with each other and are
commonly used in existing literature on traffic studies (Daganzo et al., 1999;
Tong, 2005; Varaiya, 2005). There are many kinds of “delay” that can be
defined and measured. Similarly “congestion” has many interpretations. For
some “point” facilities, such as intersections, average delay is the defined level
of service indicator. For arterial streets, the average travel speed of the vehicle
is the defined indicator, although the data is often taken by means of travel
time run. Even for freeways, where density is the defined level of service
indicator, speed, delays and related travel times are also critical measures
when describing congestion. Travel time has the advantage that successive trip
segment can be added directly, without concerns for weighted averages.
Speeds, on the other hand, do not have to be linked so directly no trip length
in the presentation, and are often used to describe performance at a particular
point or short segment of a facility (Roess et al., 2011).
Traffic congestion has no real concrete definition, but it can be best described
as vehicle density, vehicle speed, and traffic volume. Vehicle density refers to
the size of the vehicles on the road. Vehicle speed is the rate at which the flow
of traffic is moving while traffic volume deals with how many cars are on the
road during a certain time, say, in 1 hour. There are essentially two types of
traffic congestion. It can either be recurrent or non-recurrent. Recurrent
congestion is every day rush-hour stop and go traffic, occurring when the
100
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
capacity of the freeway is exceeded. The initial cause of this type of traffic is
that there are too many people rushing along our highways usually alone in
their cars during the peak traffic back-up hours, hence the name ‘rush-hour’.
These hours range from 7 to 9 A.M. and from 4 to 6 P.M., the time when
people are traveling to and from work, respectively. This recurrent congestion
deals with the everyday drivers who usually know how to handle the problems
or situations that might occur in a traffic jam. This is an important
consideration when dealing with an efficient way of getting out of the traffic
jam. The more equipped the drivers in the traffic are with the skills and
knowledge of dealing with this situation, the easier and quicker the traffic will
dissipate (Roess et al., 2011). Non-recurrent congestion, on the other hand, is
the unexpected or unusual congestion caused by an event that was unexpected
and transient relative to other similar days. These can be caused by a variety
of factors such as lane blocking accidents and disabled vehicles, construction
lane closures, inclement weather, and significant increases in traffic volume
in comparison to “normal” traffic volumes.
101
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
2. Methodology
2.1 Preliminaries
If each class of driver drives at its desired, “free-flow” speed, the observed
uncongested flow-density relationship is a weighted average of the desired
speeds. Assuming such behavior, a “i” yields estimates of free-flow speeds
for these classes. Under uncongested conditions, the total uncongested flow
𝑞𝑣 is given by
𝑞𝑣 = ∑𝑖 𝑣𝑓𝑟𝑒𝑒,𝑖 𝑝𝑖 𝑘 (1)
where 𝑣𝑓𝑟𝑒𝑒,𝑖 is the free-flow speed for each vehicle class i, 𝑝𝑖 𝑘 is the
density of vehicle class i per unit length of roadway, pi is the proportion of
1
vehicles of vehicle class i, 𝑘=∑ , 𝑠𝑖 is the inter-vehicle spacing (front
𝑖 𝑝𝑖 𝑠𝑖
to front) (Kockelman, 2001). However, under congested conditions, the
driving situation becomes very different. Speeds are no longer constant for
rising densities, and drivers can no longer choose their free-flow speeds.
Instead, each class of driver is able to control the spacing at which it follows
the preceding traveler. One behavioral assumption used in various studies is
that spacing 𝑠𝑖 for each driver class i is a linear function of speed (Daganzo,
1999).
In the Philippines, public utility jeepneys (PUJs) are the most common means
of public transportation. According to the Land Transportation Office (LTO)
Administrative Order (AO-2009-018), the standard size of a PUJ in the
Philippines must be 11m in length, 4m in height and 2.5m in width. In this
study, flow rate of vehicles is set at an average of 10 km/hour as observed
during peak hours.
Currently there are 23 PUJ routes that pass through the CDB of Cagayan de
Oro City. The southbound PUJs are usually not allowed to enter the CBD,
hence are not included in this study. There are 4 routes for Eastbound PUJs,
namely Bugo/Puerto, Lapasan, and Gusa/Cugman (RC and RD). Eleven
102
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
routes for Westbound PUJs, namely NHA (RA and RB), Patag (C2), Apovel,
Carmen (R1 and R2), Iponan (RA and RB), and Bulua (B1, B2, and B3) and
8 for Northbound PUJs which include Pier (RA and RB), Kauswagan (RA and
RB), Bayabas (RA and RB), and Bonbon (RA and RB). The total number of
PUJ units operating on these PUJ routes is shown in Table 1.
Table 1. Total number of PUJ units operating in the Eastbound, Westbound and
Northbound routes of Cagayan de Oro.
Table 2 shows the actual average traffic count of PUJs in various streets
around the CBD of Cagayan de Oro obtained during peak hours (7-9 AM and
4-6PM) during the month of June 2013. Based on actual observation of the
real traffic condition of Cagayan de Oro’s traffic network and the data
obtained from RTA, congestion mostly occurs in C.M. Recto Avenue,
Yacapin Extension, Vamenta Boulevard, J.R. Borja Street, Capt. Vicente Roa
Street, Ysalina Bridge, Don A. Velez Street, Marcos Bridge, Sergio Osmeña
103
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
Table 2. Actual average traffic count of some streets in Cagayan de Oro during peak
hours.
Street, Gaabucayan Street and Castro Street. Yacapin Extension and Capt.
Vicente Roa Extension are narrow roads yet most of the PUJs are required to
pass through these streets since it provides access to Cogon Public Market.
Most PUJs are also required to pass through Castro Street, which provides
access to Carmen market. Moreover, most eastbound PUJs and some
westbound PUJs are required to pass along the streets passing through the
Market City in Agora.
104
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
In order to determine the new routes for the traffic network around the CBD
of Cagayan de Oro, critical links were identified. A link is considered
“critical” if more than 5 PUJ liners pass along this link and its total link flow
exceeds the maximum link capacity. Based on the actual observation
conducted and from the information obtained from the RTA, the critical links
in the road network around the CBD of Cagayan de Oro were identified. The
attributes of these critical links, based on the data provided by the RTA, are
shown in Table 3. The link capacity of each link is computed by taking into
consideration the occupancy value of each PUJ and the assumed flow rate of
vehicles in the network during peak hours.
Based on the attributes of the traffic network around the CBD of Cagayan de
Oro, a maximum flow problem (MFP) was then formulated to identify how
much traffic flow should be allowed on specific streets given the link
capacities of these streets. This formulation is as follows:
105
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
where each 𝑥𝑖𝑗 is the decision variable designating the flow passing through
each link (i,j) from source node 𝑖 ∈ 𝐼 to node 𝑗 ∈ 𝐽 and L is the set of all
links in the network being considered. The MFP (1) aims to send as much
flow as possible between two special nodes, a source node i and a sink node j,
without exceeding the capacity of any link in the network. Hence, the
objective function z in (1) is the total number of PUJs allowed to pass into the
sink nodes without exceeding the link capacity uij. Note that in this
formulation, only the links identified as critical links as shown in Table 3 have
limited link capacities. It should be noted that not only PUJs can pass through
these critical links. However, this study only considers PUJs as the main users
of the critical links since there are various alternate links available for the other
modes of transportation. The traffic network surrounding the CBD of
Cagayan de Oro is shown in Figure 1. Each of the links in this network was
considered in the MFP formulation (1).
By solving the MFP (1), the maximum number of PUJs that must be allowed
to pass through the links around the CBD of Cagayan de Oro were determined.
In particular, the solutions of MFP (1) provide the total number of PUJs that
must be allowed to pass through the critical links around the CBD of Cagayan
de Oro. Table 4 shows the maximum number of PUJs allowed to pass through
the critical links around the CBD of Cagayan de Oro as compared to the
current number of PUJs passing through the critical links.
It can be seen from Table 4 that based on the current routing scheme being
implemented by the RTA, there are critical links in which the total number of
PUJs passing through these links exceed the maximum capacities of these
106
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
links (e.g., Corrales street and S. Osmeña Street). However, after solving MFP
(1), link capacities of the critical links have been strictly observed and hence,
the total number of PUJs allowed to pass through each link in the network
around the CBD do not exceed the allowable link capacities. Taking into
consideration the maximum number of PUJs allowed in the critical links,
proposed routes are suggested and are shown in Figures 2-8.
It should be noted that these proposed new routes only considered the link
capacities of the links in the network surrounding the CBD of Cagayan de Oro.
107
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
Table 4. The proposed maximum number of PUJs that must be allowed to travel in the
critical links as compared to the existing routing scheme.
Capt.
570 9 441 3 562
Vicente Roa
Corrales
130 6 418 5 112
Street
The MFP formulation (1) does not consider the effect of the proposed
rerouting to the riding public. Some of the proposed new routes may cause
inconvenience to the PUJ commuters but these new routes are expected to ease
the currently congested critical links in the network.
108
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
109
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
Figure 7. Proposed routes for Pier Macabalan RA, Pier Macabalan RB and Bonbon
RA
110
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
In this study, a model for optimizing link capacities in the traffic network
surrounding the CBD of Cagayan de Oro City was formulated and solved
using maximal flow problem formulation. Solutions to this MFP provide the
total number of PUJs that should be allowed to pass through each link in the
network. By allowing only a specified number of PUJs to pass through these
critical links, traffic congestion around the CDB of Cagayan de Oro may be
minimized.
This study only focused on the public utility jeepneys plying through the
streets around the CBD of Cagayan de Oro. For future studies, it may be
interesting to include other vehicles to provide a more realistic approach to
alleviating congestion
5. References
111
R P. Agdeppa-Namoco / Mindanao Journal of Sci. and Tech. Vol. 11 (2013) 99-112
Li, X., Han, J., Lee, J.-G. and Gonzalez H., (2007). Traffic Density-Based Discovery
of Hot Routes in Road Networks, Advances in Spatial and Temporal Databases Lecture
Notes in Computer Science 4605, 441-459.
Morarescu, I.C. and De Wit, C.C., (2011). Highway traffic model-based density
estimation. In Proceedings of the American Control Conference 2011, San Francisco,
USA.
Roess, R.P., Prassas, E.S. and McShane, W.R., (2011). Traffic Engineering (4th Ed).,
Prentice Hall, USA.
Tong, L., (2005). Nonlinear dynamics of traffic jams, Phys. D, 207, 41–51.
112