DHSUD’s New TOD Guidelines: Public Transport Realities, Data Gaps, and Technical Refinements
DHSUD's new guidelines on mainstreaming transit-oriented development (TOD) into local land use plans and zoning ordinances are a milestone. Here, we wish to present some ideas, offered in a spirit of collaboration, for making the complementing technical tools and guidelines, or next edition even more useful to planners on the ground.
Last October 21, 2026, The Department of Human Settlements and Urban Development (DHSUD) launched to the public the new Guidelines on Mainstreaming Transit-Oriented Development (TOD) into Comprehensive Land Use Plan (CLUP) and Zoning Ordinance (ZO). This was part of the GIZ-supported Urban Act project handover event.
As highlighted in the excerpt below, the Institutional Arrangement policy embeded in the guidelines emphasizes TOD as a central tool in urban planning. Therefore, to put this policy into practice, planners in both national agencies and LGUs need to be equipped with TOD tools and more specific know-hows.
All relevant agencies are required to align their land-use and transit strategies under this central body’s framework to eliminate planning silos, prevent delays in critical infrastructure, and ensure a consistent, integrated approach to urban development. (TOD Guidelines, pg. 121)
Bringing Transit-Oriented Development (TOD) into the local Land Use planning process is a great way to re-calibrate how Philippine cities evolve, and DHSUD, with its development partners including GIZ, deserves credit for it.
The TOD Guidelines (2026) weave TOD into the 12-step CLUP process, offer a station typology, propose a site-selection scoring system, and connect TOD with affordable housing, inclusion and climate resilience. Planners from both the government and private practice have wanted this kind of common reference for years.
At SafeTravelPH, we work on local public transport, so we read all 183 pages with the eyes of people who will try to apply them. (The rest of the clerical stuff can be simply done to your AI-powered checker.)
In this post, we share what stood out to us, starting with the evidence behind transit planning. We look at how the Local Public Transport Route Plan (LPTRP) is used as a source, and how public utility vehicle (PUV) supply and demand could be estimated more robustly. We then show how isochrones make walkable catchments more realistic than simple rings, using a local example. Next, we look at how the number of transport modes and services, including where their routes overlap, can act as connective fibre for a transit hub.
We also argue that intersections will be among the most coveted places for neighborhood and transit-core innovation, and look at how they could be redesigned. Finally, we turn to managing private vehicles, asking whether parking requirements make sense in our TOD vision. Throughout, we treat density as activity and interaction per unit of place, not just people per square kilometer.
The DHSUD TOD Policy Framework aimed at promoting ‘efficient and sustainable land use development around emerging public transport terminals and stations’ to ‘create a wide range of business, housing, civic, and leisure activities. Based on the policy, “TOD can be defined as a development within 1 kilometer from a transit station that provides sufficient densities, mix of activities, and pedestrian links to support significant transit ridership”. NHSB Resolution No. 2023-02 “Adopting the Transit-Oriented Development Policy Framework” (Defining TOD, pg. 19)
The TOD Basics: Zones, Housing, Places, Public Transport
The guidelines aptly cited Ann et al (2019) research on TOD catchments for setting the baseline distance, as examined in the paper for the case of India transit stations. It starts from how far people are willing to walk to reach transit, usually somewhere between 400 m and 1,200 m. The guidelines deliberately stretch the influence area beyond that range so that affordable housing can be brought inside it.
They also remind planners that although TOD is carried out parcel by parcel and street by street around a station, it should be planned across several connected scales, with the influence area as the concept that helps us understand its wider spatial impact. This is a sensible idea, but it has a consequence. Once housing sits beyond comfortable walking distance, residents depend on feeder services to reach the station.
That is why the typology tables, which call for “high accessibility for PUVs” in urban and suburban centers and neighborhoods, deserve a closer look.
For example, one concern is that “high accessibility” is stated as a suburban outcome without a standard behind it, and we know from experience that reality often falls short (TOD Station Typology Characteristics, pg 31). Research on city outskirts shows that PUV service thins out sharply there, especially during off-peak, and that a suburb is often served by a single line. Where a corridor does have frequent service, it is often because routes from neighboring LGUs happen to pass through it, which no suburban plan counts on or controls. This implies that supporting PT supply to TOD zones must consider even metro-wide transit planning and governance. For further illustration, we will share how the case of a future NSCR train station in Guiguinto, Bulacan LGU shows a related issue connecting TOD work and existing PT supply reality.
Relatedly, as the guidelines remind us, the goal of TOD is prevent the economic and environmental toll of urban sprawl:
At its core, TOD promotes compact city growth by integrating land development and transport services, which counters urban sprawl. (Mainstreaming TOD in CLUP and Zoning Ordiannces, pg. 57)
Making TOD Work for the Jeepney, the Bus and the Crossing: Public Transport-based Reflections on the New DHSUD TOD Guidelines
The guidelines also set the clear policy that TOD should not be limited to anchoring developments around rail transport hubs, as we know that Philippine cities outside the Greater Capital Region relies heavily on a less-simpler, less predictable road-based public transport system.
It is also possible for LGUs to plan and implement TOD-related projects, especially for non-rail transport. For instance, many LGUs have transport terminals or hubs that can have different transport modes and may be located in government-owned lands or private lands with special arrangements with the private sector. LGUs may also undertake public-private partnership (PPP) arrangements for projects that include terminal hubs and local commercial establishments. For PPP arrangements, it is essential to clearly define the role of the private sector. These include investment in community development, infrastructure construction, building management, and operation of completed facilities. (Integrating Land Use and Transportation Planning, pg. 7)
While the DHUSD guidelines establish a strong framework, a few targeted refinements could make the tools even more effective in practice. We offer these observations from a public transport planning and operations perspective to help sharpen the conversation. We will take a closer look at each of these topics in future posts, with more examples to help government and private planners put them into practice. For now, here are the issues we think matter most.
Density is Destiny
The site-selection rubric in Annex C may be the most practical tool in the document, which is why it deserves the most attention.
The top population-density score goes to areas below 1,500 people per square kilometer. That is well under what frequent transit typically needs, and far below built-up Metro Manila, so the rubric can't yet tell a dense station area from a modest one. This is where the idea of transit-supportive density could help: minimum densities tied to the level of service a hub is expected to provide.
Meanwhile, in the Philippines, city-level densities are heavily right-skewed. Across the 149 cities in the 2024 POPCEN data, the median density is about 985 persons/km² (about 4 person/acre), meaning half of all cities fall below this mark. The middle 50% of cities span from approximately 460 to 4,000 persons/km², and the outliers extend from 133 (Puerto Princesa) up to an upper statistical outlier fence of about 9,300 persons/km². Nearly two-thirds of cities (97, or 65%) maintain densities below 2,000 persons/km², only 35 (23.5%) reach 5,000 or more, and just 20 (13.4%) reach 10,000 or more. Natural reference points within the distribution are roughly 1,500 (the top of a typical component city), 6,000 (the top quartile of highly urbanized cities outside Metro Manila), and about 9,300 (the statistical outlier fence).
By classification, only the highly urbanized cities (HUCs) are dense as the norm (median 10,450), a figure driven entirely by Metro Manila's compact geography. The 17 HUCs outside the NCR operate on a different scale, with a median of 1,797 and a top quartile starting at roughly 6,050. Component cities have a median of 723, with 16 statistical outliers. These outliers are primarily concentrated in the suburban commuter corridors of Cavite, Laguna, and Bulacan, featuring extreme densities like Bacoor (14,325) and Biñan (13,436) that eclipse most HUCs. The five independent component cities fall in between, with a median of 2,178 and the highest density recorded in Dagupan at 4,695.
One caveat: these are gross citywide averages that calculate 2024 population against officially certified land areas, meaning the densities of expansive territorial cities are significantly diluted by farmland, mountains, and water, and do not reflect the true station-area densities of their built-up downtown cores.
That is also why they work as a floor. City centers are always denser than their citywide average, so a city whose average already clears a benchmark has cores that clear it comfortably, while a city below it may still have TOD-viable corridors. So we can read them as conservative lower bounds for comparing cities and setting minimum screens.
The guidelines already list "alignment of densities and transit capacity" as a principle, and the next step is to attach numbers to it. Meanwhile, the proposed site selection scoring of 40%+ commercial land as the best rank may pull against the mixed-use and housing goals the document sets elsewhere. To counter this, the Transit-Supportive Densities research also provide guidance on including work/employment with the residential density:
Isochrones: The Transport Planners's Best Friend
In the guidelines, distances are drawn as circles of 400 m to 1 km (up to 5 km for support areas), but the terms "isochrone" and "walkshed" don't appear in it. A circle can't see an expressway, a river, a long signal wait or a steep slope. Time-based network catchments handle all of these, and free routing tools make them within reach of any LGU.
Unlike the fixed-radius circles commonly used to define station catchment areas, isochrones map the places people can actually reach within a specified travel time, following the available street network. In many Philippine cases, streets are disconnected, pedestrian routes are obstructed, or physical barriers make a station that looks close on a map difficult to reach on foot. A one-kilometer radius may establish that a location is near a transit station geographically, but it does not establish that the station is accessible in practice.
Two recent publications offer a stronger basis for incorporating this approach into TOD planning. In Planning for Transit-Oriented Development in Emerging Cities, Arturo Ardila-Gomez and co-authors (World Bank, 2024) discuss isochrone accessibility as a way to measure the number of opportunities, such as jobs, that people can reach within a specified travel time using different transport modes.
Meanwhile, Abdul Halim, Teh, and Rosni (2026), in GIS-Based Walking Map Isochrones in Measuring Transit Catchment: A Spatial Approach to Urban Accessibility, demonstrate how walking-network catchments can differ substantially from conventional circular buffers around rail stations. For the three stations studied, the 5-minute walking catchments were 16.3% to 60.8% smaller than their corresponding 400-metre circular buffers. The 10-minute catchments were 28.2% to 65.3% smaller than the corresponding 800-metre buffers. The differences varied by station, demonstrating that a simple radius does not reliably represent the area people can reach on foot.
Together, these studies point to a practical improvement for Philippine TOD guidelines: assess not only how much development lies within a designated radius, but also how easily people can reach transit and the opportunities it connects them to. It also provides a good picture of the threats in implementing parcels and streets-based development set out by the TOD policy, as mentioned earlier.
More on Isochrones + Road Transport
To further illustrate with a local example and data, we are sharing our own data for NSCR station in Guiguinto, Bulacan. An initial catchment work drew the standard concentric rings (200 m, 800 m and 1,000 m) around the station, and those circles run straight across the NLEX access road and interchanges, and the rail line.
Meanwhile, the 5-, 10- and 15-minute walking isochrones tell a different story: the walkable area is lopsided, following the street network and stretching along the rail corridor. The five-minute zone is largely open land next to the assumed station entrance (left/west of the station), the denser neighborhoods begin near the 10-to-15-minute edge, and the dense housing across the expressway access/spur road to the northeast appears to fall outside the 15-minute reach altogether. (Download the tiny html app of the Guiguinto NSCR Station Isochrone here. Just load the html file in your web browser.)
The same station area shows why the road-based transit supply needs as much attention as walkability. Our data on surveyed jeepney routes, 2.8 to 5.6 km long, took 16 to 32 minutes, with an average speed of about 10.5 to 13.1 km/h. That is roughly half the 20–25 km/h planning value DOTr suggested for LPTRP's fleet sizing and dispatching formula, and similar to crowdsourced PUV speeds in other cities and in the far more congested Metro Manila (8–13 km/h).
Route | Distance | Average Travel Time (6–8 round trips) | Average Travel Speed |
Plaridel to project site | 3.6 km | 21 mins | 12.0 kph |
Brgy. Malis, Guiguinto to site | 2.8 km | 16 mins | 10.5 kph |
Balagtas town to site | 4.6 km | 21 mins | 13.1 kph |
Bulakan town to site | 5.6 km | 32 mins | 10.5 kph |
The delays are mainly due to waiting and frequent stops for boarding and alighting, clustered at the C. Mercado and MacArthur Highway intersection, which is about 2.3 km (a 30-minute walk) from the station. The survey was small and held at mid-morning, so it is indicative rather than definitive. Still, it suggests the local transport plans should be revisited as a source for TOD metrics, as prescribed in the guidelines, and refreshed with measured data.
It also suggests feeder services should be built to high quality before the station opens, through formal stops, enforced loading and coordinated routes. Reliable feeders and rail would then reinforce each other and build the habit of choosing public transport. Isochrones can guide that work by showing which stops and street links fall within walking reach and where feeder service has to take over.
Ideally, the local transport plan for these types of LGUs, where new stations are being constructed, should be redesigned with TOD as the central focus, vis-a-vis multi-LGU, provincial, or metro-wide planning of road-based public transport services. However, this requires additional technical and institutional efforts that DOTr may need to be empowered to address. (Practical planning question: What are the plans of the government for existing bus and jeepney routes whose ridership will be affected once the new rail line is opened?)
From counting routes to understanding demand
The guidelines don't yet describe how to estimate transit demand. There is no origin-destination (OD)-based modeling or trip generation by land use, and no forecast method for planned hubs. Perhaps this would be DOTr's role in the succeeding complementary agency-initiated technical guidance notes.
The "passenger volume" criterion demonstrates why this is crucial too. It scores a percentage "at peak hours," which can be read as a peak to non-peak ratio, although a technical note justifies it through load factors. If read as a ratio, a city with a very peaky system, as many secondary metros have, would outscore Metro Manila's flatter pattern, which isn't what we'd want to reward. The ratio can also exceed 100% while the top band stops at 70%.
The TOD guidelines also refer only to boarding. Alighting or unloading activities doesn't appear anywhere. Yet the crowds at stations and stops are made of people getting off, transferring and crossing as well as people waiting. Platforms, stairs, crosswalks, curbs and waiting areas should be sized from boarding, alighting and transfer flows by time period.
These flows can be estimated from operator and app data, counts and OD matrices, or modeled in tools like Cube or MATSim. A published study of Puerto Princesa's route plan has already applied Cube Voyager network modeling, so the capability exists locally.
Building on the LPTRP, with a little help
LPTRPs are named as the data source for ridership, feeder modes, frequency and housing access, and they also determine which hubs become TOD sites. That's a lot of weight for one document, and preparing LPTRPs is genuinely hard.
As our review of the LPTRP framework shows, the current government guidelines on route assignment of PUVs, demand estimation, and fleet-sizing would fail in addressing both the commuters service requriements and the operator's viability to maintain a quality level of service.
None of this is a criticism of LGUs. It reflects limited data and tools. Because an LPTRP describes planned supply, not demand, we'd suggest pairing it with independent checks such as boarding counts, OD surveys and network modeling. Otherwise any weakness in the plan carries through to hub selection, zoning, and conflicts among the road-based transport providers and users.
Connectivity by service, not just by mode
The connectivity criterion counts feeder modes, so seven jeepney routes converging on a station count as a single mode, while a tricycle line, a pedicab line and an e-trike line count as three. The monitoring table counts routes instead, so the two sections would benefit from alignment.
More routes are usually a good sign, and overlapping routes on one corridor create both redundancy and connectivity. But bunching of vehicles on limited space near the transit hubs/stations are a common sight in Philippine roads. Measuring combined frequency, capacity and how many distinct origin areas are reached would give a truer picture.
Where the buses and jeepneys will go
So if we could add one thing to the next editions, design-wise, it would be space for land-based transit. The guidelines cap parking and mention "structured terminals," but don't yet set standards for PUV bays, holding areas, dwell space, circulation or curb length. Anyone who has watched buses and jeepneys bunch up around a Manila rail station knows this is something we should correct in our infrastructure. Bays can be sized from headway and dwell time, and larger developments could provide off-street terminal space.
Three related topics deserve a place in the next guidelines.
Intersections. Big intersections often have the land uses and buildings best placed for TOD, so they'll be coveted parcels. Today they're typically designed for vehicle flow, with PUV loading discouraged and pedestrians sent over footbridges. The guidelines touch on intersections only for crosswalks, intersection density and traffic-impact review. Helpful additions would include an intersection and user priority typology, a right-of-way priority protocol, and a joint planning-engineering review so that planners and traffic engineers work from the same playbook. Traffic assessments could also measure people moved, not only vehicle flow.
Bus terminals outside rail areas. Most of the country won't have rail soon, so bus terminals will be the anchors. The guidelines list them as potential hubs, and the next step is defining what makes one high-quality, with a provincial-level hierarchy in the route plans. There's a practical constraint too: a government-built hub can't simply require existing operators to move their private terminals according to a Supreme Court ruling. Route-plan and franchise conditions, incentives, and joint ventures with operators are the levers worth spelling out.
Mall terminals. Malls that run their own terminals and charge per dispatch aren't mentioned, yet where they sit on suburban, car-oriented sites (usually, for secondary cities) they're de facto hubs. Planners could inventory them and either designate them as TOD nodes with conditions (public access, permeability, reduced parking) or at least account for the ridership they capture.
What's next?
Bridging the gap between a 183-page policy milestone and functional streets will not happen through land-use zoning alone. For the DHSUD guidelines to reshape Philippine cities rather than sit as aspirational CLUP annexes, technical and institutional follow-through must happen across three fronts:
A Joint Technical Guidance Note with DOTr and DILG. While DHSUD has laid the spatial foundation, the DOTr and LTFRB must supply the transit operations playbook. A supplemental technical bulletin should update LPTRP manual and/or its formulas to reflect real-world operating conditions and business and financing requirements, improve intersections policy and management, establish clear geometric sizing standards for on- and off-street jeepney and bus bays to prevent curb bunching, and provide policy mechanisms to integrate private commercial terminals into the formal TOD hierarchy.
Equipping LGUs with Low-Barrier, Tiny Tools. Local planners do not need expensive proprietary software to make TOD metrics real; national agencies and development partners should instead champion standard, open-source toolkits. The work to identify and prepare a TOD may be cumbersome at a glance. However, with advancements in AI, local planners can now develop their own small applications (like the isochrone HTML app we shared) to enhance their understanding of their local areas and their potential. Beyond planning, the management and governance of these developments would necessitate not another software system, but a group of capable policy movers.
Rethinking Intersections as People-Movers. Local zoning must bridge the divide between spatial planners and traffic engineers so that intersections within TOD zones are not optimized solely for motorized vehicle flow. By adopting street design standards that prioritize direct at-grade pedestrian crossings, dedicated PUV loading spaces, and multi-modal assessments that measure people moved per hour instead of vehicles cleared per cycle, we can ensure a TOD feels like a walkable neighborhood rather than an isolated transit island. We must avoid the past mistakes seen at rail stations in Metro Manila.
Stay tuned in our upcoming posts as we try to release more knowledge products and discuss practical ways to develop capacities in local and national agencies. We look forward to unpacking these tools in detail with you.
