Project Everest

Proposed Experiment

Democratise waste collection

Ciaran Hoare
Ciaran Hoare | Jul 11, 2017 | in Knowledge Base

Perhaps a quick way to find out if someone else has solved an issue you are trying to solve is to google “the Uber of X”. Maybe I’ll google the Uber of friends ¯\_()_/¯.

In combining the previously posted ideas of RFID tags and the Wings payment system, there exists an opportunity to keep most of the associated costs of ERS variable and reduce overheads.

This idea basically consists of a system where ‘jobs’ would be lodged through an online/app/text system in order for those wanting to make some extra money throughout the day to pick up and dispose of at the appropriate location. The image attached below offers a poorly drawn flowchart for the logistics of this type of system.

This system holds everyone accountable for honesty as the number of bags picked up can be cross referenced between the household and the driver, and the number of bags dropped at each depot can be cross referenced between the driver and the depot.

Drivers could be paid in a similar style to how UberEats operates, scaled appropriately to the country in context. Currently UberEats bills a total of $6.80 per pickup, $2.20 for drop off, and $0.95 per km ridden, of which the “partner” gets 65% and Uber gets 35%.

This could also have applicability in Timor through the yellow taxis, and would mitigate the negative impact of reducing income of the boys who push the carts around and collect cans, as they could be onboard as collectors as well. (Though trucks would still be required for Timor Plaza etc. this could serve as a means to impact smaller communities).

In short: Greater supply of materials for depots and recycling companies, additional income for transport workers when they’re not driving people around, easy way for households to get their rubbish taken.

A potential drawback of this system is the movement and exchange of the waste collection bags that would be supplied to households and communities. How could this be made a closed system to minimise resources needed to return empty bags to communities and to minimise lost bags? It’s success is also hinged upon financials and the technological requirements for different stakeholders involved.

Here is a similar concept applied in Chile:

edited on 11th July 2017, 14:07 by Ciaran Hoare

Chris Swanepoel Jul 11, 2017

I posted a similar idea from the ERS Cambodia team tonight - this is a really exciting avenue! We're looking at closing the system by having the request for a bag to be picked up come into the system once the contents of the bag reaches a level marked on the bag. This will be followed by an automated reply text or prompt in an app to request new bags. The room above this line would give a time window during which the collection needs to happen, giving time for a viable route to geographically close customers to be calculated. The bags would then be swapped out by the driver and the new unique codes noted or scanned, keeping them traceable to a connected phone number or app account. This way a bag collection and delivery trip will only be done if it is at maximum capacity and within reasonable time constraints, ideally limiting variable costs of sending out drivers on a regular route and not collecting at full capacity.

Reply 6

Daniel Shea Jul 12, 2017

Consideration for the texting system to exist in countries where online banking and smart phone technology. The use of payment back and forth with phone credit is a norm and understood act within places such as Timor. Using a text notification system for collection, retrieval.

Reply 2

Daniel Shea Jul 12, 2017

And have the check system at the dump to confirm and post process the owed credit to the initial parties. payment for delivery lies in the split you discussed.

Reply 1

(Account removed) Jul 12, 2017

The idea of using a Uber based model for collecting waste definitely has potential.

My major concern is that the cost of collection per bag would remain, as a whole, reasonably high. On the assumption that you are basing this idea of the Uber model, where drivers respond fairly quick to orders, my concern is that the trips made by drivers won't be optimised. Instead of potentially collecting multiple bags from a community in one trip, lowering transport costs, bags would be collected based on when the households are ready. This is not necessarily cost effective, as the cost of collection per bag would be higher.

To mitigate this, drivers could respond to to trips only once a certain number of bags in a specified area are ready for collection, instead of having multiple drivers within days of each other collecting from one region.

Reply 3

Ciaran Hoare Jul 12, 2017

Net profit = X * (Plastic price) + Y * (Glass price) + Z * (Cardboard price) - A * (Cost per pickup) - B * (Cost per drop off) - C * (cost per km)
Where x,y,x,a,b are no. of bags an c is km driven.
Subject to constraints:
Threshold * Vehicle capacity < X+Y+Z < Vehicle capacity
(Gross profit - net profit) / (Gross profit) > Margin threshold

Reply 0

Kai Faulkner Jul 12, 2017

I think the notion of 'Uber-fying' this process to reduce overheads could be slightly off the money.
It is common knowledge that Uber is one of the fastest growing businesses ever being now valued somewhere around the $70billion mark. However it still runs at a loss due to administration costs and other operational costs.
They manage fine (maybe even thrive) due to the sheer scale of their operation but for a smaller business still searching for customer validation with a bare-bones MVP this would be like cancer.
So my question, what is wrong with the conventional waste removal system that we see in countries such as Australia. The operation im visualising looks very simple. (see attached image)

Reply 1

Ciaran Hoare Jul 12, 2017

Fair point, the whole idea is entirely hinged upon what the $$ do.
Scalability and overheads (trucks and property) are the drawbacks of the simplified model

Reply 1

Kai Faulkner Jul 12, 2017

yeh of course, i think the key is going to be minimising risk whilst increasing efficiency, scalability and $$$.

Reply 1

Chris Swanepoel Jul 12, 2017

In our Cambodian case the main draw backs of the simple weekly collection model are:
- Not necessarily every house will be serviced at this stage since we cannot make it like a mandatory part of Council rates as in Australia
- The disparity in income levels serviced means not every bag in every house will be filled each week e.g. glass and cans will fill at irregular rates meaning the service won't accomodate for the needs of different households and there is a greater chance of running way below capacity with high costs.
- The high capacity of standard garbage trucks makes it unviable to collect few jobs, however the limitations of tuk tuk trailers means small collections need to happen anyway so might as well fill the trailer to capacity whenever possible

These points can be overcome by changing cycles depending on how quickly things fill up but the risk of showing up to a house and the owner saying "No, don't need it collected yet" is too high and asking a fee for taking a quarter bag seems unreasonable given the low disposable income.

Definitely in the mean time while the business is set up the weekly collection model fits the MVP best however our as our financials are indicating already, the asking price will need to be very high to cover costs.

Have a read of our system flowchart attached if you didn't see it on the other post.

We are aware that there are still a number of holes that need considering and really appreciate an outside perspective in pointing them out, that's why we're here!

Reply 2

Kai Faulkner Jul 12, 2017

Awesome Chris, absolutely valid points, and so true, you guys are are there, and i haven't been fortunate enough to travel to cambodia yet so i'm glad you really are making this focused on the users. Maybe you could run some experiments to test out those assumptions you have made though?
My concern still stands though that there are a lot of moving parts there in that flow chart. Again back to the uber analogy, its like the uber vs taxi argument, yes overall Uber is a better experience tailored to the user, but the taxi is there always when youre walking the street, just stick out your hand and you have a ride.

Reply 1

View all replies (4)

Darcy Connaghan Jul 2, 2018

Status label added: Proposed Experiment

Reply 0