Modular Community-Driven Housing for Tondo, Manila

What if affordable housing could do more than shelter? This “Make It Home” project began with that simple question, but quickly grew into a larger conversation about authorship, adaptability, and the ways that space and community can support life.

My name is Natalie Lo, and I’m proposing a modular housing system for the Tondo District in Manila, where the patterns of the place and the people will shape our landscape. Our system is derived from the concept of incremental housing, where a core “chassis” unit, identical from house to house, forms the basis from which residents can select additional housing modules. I am prioritising evolution through participation - residents aren’t just recipients of housing, but they become active contributors in this flexible yet rooted system.

Our idea draws from the incremental housing of architect Alejandro Aravena’s Elemental project in Chile, where half-houses were built with the other half left for residents to design. It also follows the thinking of Jane Jacobs, who resisted large-scale modernist transformations of cities from a top-down design perspective, and instead prioritised the wisdom of lived-in streets. Both Aravena and Jacobs offer lessons in how not to over-determine design; instead of planning everything from above, this model opens space for communities to shape what they need - collectively and in their own time.

The image above is the Elemental’s half-houses in Chile.

Supplies I used include Fusion 360, cardboard, hot glue, paper, and popsicle sticks.

Tondo is one of the most densely populated places in the world, and much of its built environment has developed outside formal channels. This makes the urban landscape highly precarious - poor access to infrastructure, with vulnerability to intense climate. However, this is also what makes Tondo deeply adaptive, and the perfect place to implement this housing model.

Rather than approaching this context as a problem to eradicate and fix, our project tries to work with what’s already functioning. Informal settlements often contain sophisticated social and economic networks - with neighbors depending on each other on a day-to-day resource basis, homes doubling as workspaces, streets being used for everything from childcare to commerce. It was from this analysis that I noted the importance of incorporating communal facilities into our design. Not only should our housing framework provide private spaces, but also facilitate communal interaction. In that sense, Tondo doesn’t lack design; it lacks support.

The redevelopment of Dharavi in Mumbai - a similarly populated neighborhood filled with informal settlements - offers a cautionary parallel. There, a top-down plan has been criticized for pushing infrastructure without regard for the social and economic lives already taking place on the ground. The concern is not just displacement in the physical sense, but the erasure of a working system. This project tries to respond differently: by offering housing that strengthens what’s already present, without flattening it.

I looked at many similar examples in history of modular housing to see what I could learn. The Khrushchyovkas built in the Soviet Union, for instance, were conceived as an efficient solution to a housing shortage. They were cheap and fast, as they were easy to replicate - but also isolating and rigid. Over time, their uniformity became a liability. They couldn’t accommodate change, nor could they support the kinds of informal social life that make dense urban neighborhoods livable. From this, I was reminded that speed and scale alone aren’t enough.

The first image is of the Tondo District, Philippines. The second image is of Khrushchyovska

I started with a chassis - a core unit that could be lived in immediately, but that still had the potential to adapt. I wasn’t interested in designing a finished house. I wanted to offer a starting point - something that could support people’s needs now, and keep supporting them as those needs changed. The chassis repeats from household to household, but it’s not meant to feel uniform. It’s just a base - something sturdy and consistent that leaves space for difference.

The core includes the foundation, roof, plumbing, electrical setup, and structural walls. Inside, there’s a single living space with a window for ventilation, overhead lighting, and electrical outlets. At one end, a compact wet core contains a shower and toilet, located to minimize plumbing complexity and leave the rest of the floor open.

Each unit comes with two modular slots - one for private expansion, one for communal use. These modules are where variation happens. The core is shared, but what grows out of it reflects the specific rhythms and needs of the people living there. The level of modularity changed a few times while I was developing the idea, but these drawings reflect the version that felt most balanced - a standard chassis, paired with modules shaped by individual and collective choice. The images are of early concept drawings exploring this idea of a resident-selected vs a communal module.

Module A sits above the chassis on the second floor. It’s designed for private expansion, but the specifics of what gets built are left open. There’s technical guidance available, and some subsidy for materials, but the idea is for residents to shape the module according to their own needs. In Tondo, families live in a wide range of arrangements - many include multiple generations or extended relatives sharing space. Others use their homes for small-scale work or informal trade. I wanted this module to leave space for that variety, rather than trying to define a standard program in advance.

One version of the module might be a residential expansion. In many cases, this could mean adding a sleeping area for a growing family, or a quiet space for an elder relative. I used the proportions of the banig - the woven mat traditionally used for sleeping and sitting - as a kind of informal module size. That reference helped me think about space less as fixed rooms and more as zones that respond to activity, shifting as needed.

Another version is a kitchen expansion. In many households, cooking isn’t just for daily meals - it’s also a source of income. Food vending and home-based preparation are a big part of the informal economy in Tondo. I imagined a kitchen module that could support that: better ventilation, water-resistant flooring, cladding that handles heat and smoke. It’s not meant to replace improvised setups entirely, but to offer something safer and more supportive for families already doing this work.

The third possibility is a small business workspace. This could be for tailoring, repairs, or sari-sari supply storage. Many businesses in the Philippines are run from home, sometimes with limited space or makeshift arrangements. This module would make room for that kind of work without taking over the main living space. It includes shelving, power access, and space for tools or materials.

In all three cases, the intention is the same - to support what people are already doing, and to give them the room to do it more comfortably or safely. The structure just needs to hold space for it.

Module B is placed at ground level and selected by the community. It’s designed to be shared - something that serves the immediate cluster or block, depending on what people prioritize. I spent time thinking about how to make this feel genuinely adaptable, without overcomplicating the system or trying to predict too much.

A sari-sari store module was one of the first options that came to mind. These neighborhood stores are everywhere in Tondo, often operating out of windows or small extensions. This module offers a more formal structure - shelving, weather protection, a basic electrical connection - but keeps the logic of the sari-sari intact: small, local, and integrated into daily routines.

Another possibility is a canteen module. In many low-income neighborhoods, shared cooking already happens informally - especially when ingredients are pooled or fuel is limited. The canteen module would give that practice its own space, with a shaded cooking area, seating, and surfaces for food prep. It also reflects the idea of bayanihan - not just as emergency aid, but as an everyday way of living together.

There’s also a production module - a flexible space with workbenches, tool access, and electricity. This could be used for shared projects, informal skills training, or community maintenance. A lot of this already happens in informal ways - this just gives it clearer boundaries and more support.

Other ideas came up too: a homework space with charging outlets, a Wi-Fi node, a rainwater tap for handwashing or hydration, a first aid space, a compost garden. All of these reflect things I noticed either already happening or clearly needed. I don’t see them as upgrades or extra amenities - they’re part of what it means to live well in a place that’s dense, informal, and sometimes stretched thin.

After finalizing the 2D plans, I moved into CAD. This phase was less about getting everything polished and more about understanding how the components actually behave in space. The drawings gave me a system - this helped me test whether that system could physically hold.

I started with the chassis. It was the most straightforward: four walls, an integrated wet core, and a structural spine where modules would connect. I kept the wall height at 8 feet to stay within conventional framing and to keep material use efficient. Early on, I had placed the wet core toward the center of the back wall, but in CAD, that made circulation awkward - especially near the door. I moved it into the corner, which also freed up access to Module B and made the space feel less boxed in.

Rebalancing the Modules

Once the core was modeled, I brought in the two modules - A on top, B on the side. This required a lot more adjustment than I expected. The original design had Module B occupying a rear quadrant, but it ended up obstructing both core access and the potential for shared public frontage. Instead, I shortened its footprint and brought it forward to become a kind of anchor at the front corner. This made it more visible and easier to join with neighboring units if the communal modules are eventually linked.

Module A was simpler to place vertically, but scaling it took more time. I wanted it to feel like an expansion, not a second story stacked directly over the chassis. Resizing it slightly smaller helped keep its weight in balance with the core, and also opened up more flexibility for roof placement and stair access.

Staircase Iterations

The staircase went through a few versions. I tried interior stairs first, but they ate up too much floor area and narrowed the usable space within the chassis. I experimented with a back corner approach, but it made the façade feel disjointed and again clashed with the core.

Eventually, I settled on an L-shaped modular staircase tucked inside, behind the front façade. It’s built using a beam-and-tread assembly and repeats with an even rise, landing just before the Module A threshold. CAD helped a lot here - not just in sizing the components, but in seeing how tight or awkward certain entries would feel. With railings added, it also began to anchor the internal geometry of the home.

Roof Design

The roof took the longest to figure out. Initially, I tried a continuous sloped roof that extended from the top of Module A down to the far end of Module B. It looked elegant in sketches but proved difficult to resolve in model. The slope compromised the headroom in Module B and left a lot of awkward negative space above it.

I tried separating the roof into two planes - one for Module A, one for B - but it lost coherence. In the end, I brought back the single slope, but this time I started the ridge at a higher point and gave Module B its own flat interior ceiling. That extra vertical space could be used later for vent ducts, shelves, or even a small mezzanine for communal storage.

The slope is based loosely on the silhouette of a nipa hut, scaled out into a more modern assembly. The projection and angle help with ventilation, shade, and rain deflection, especially in monsoon conditions. I also liked how it visually stitched Module A and Module B together - not just structurally but symbolically, as two parts of the same framework.

Next, I built a cardboard model to bring my Fusion 360 design to life and better understand how the modules would fit together. The images show the different ways the modules can be assembled, starting from a single core unit. In my physical model, I built modules A and B - A on the second story, and B on the ground floor to add to the chassis - and included a small lawn to represent the communal function of module B. This was my first time creating a structure like this with cardboard, so it was definitely challenging to represent structural concepts, but it was an enjoyable process overall. Seeing the contrast between the physical model and the Fusion 360 version was interesting and gave me a new perspective on the design - it raised physical questions, such as the mechanics of how the modules would attach. It was challenging to perfectly replicate the CAD with such rudimentary materials, but many final design choices were informed by my real-life model.

This project became a way to test how modular housing might function as a flexible system rather than a fixed product. The idea was always to build something that could adapt over time - where households could start with a basic, usable core and grow it in ways that made sense to them. That felt especially relevant in places like Tondo, where many homes already evolve incrementally.

The design encourages that kind of flexibility. By separating core infrastructure from add-ons, the system keeps the upfront cost low, while still leaving room for expansion. That lowers the barrier to entry and gives families a more gradual way into permanent housing. Using local materials also felt important - both to reduce environmental impact and to keep the build process familiar and accessible.

There’s also a social dimension. Participatory planning isn’t just a method - it shifts who gets to decide what housing should look like. If residents can choose how their homes grow, or what their shared spaces are used for, it reduces waste and makes the spaces more likely to be used long term. That’s not a new idea, but it’s one that still feels under-applied.

From a climate standpoint, the design fits into broader resilience work already happening in Manila. Features like cross-ventilation, rainwater access, and raised thresholds aren’t unusual - but combining them with a modular, adaptable system could make them more effective at the neighborhood scale.

Working in CAD helped clarify proportions and connections, but it was the physical model that made things more concrete. Some layout decisions didn’t click until I started cutting and assembling by hand. Building the model helped me catch small issues - clearance around the stairs, relationships between modules - that I might have missed on screen. Moving between digital and physical formats kept the design grounded.

Overall, the process wasn’t about polishing a final product. It was about building a system that leaves space for others to shape it. That seemed like the most practical way to make housing that can last.

I'd like to acknowledge my design partner Brandon Choy for pointing me towards the Elemental project in Chile, and helping me with CAD!