Systems Engineering for Waste Management

Systems Engineering, the journal of The International Council on Systems Engineering (INCOSE) has published my most recent paper, entitled “A systems-engineering approach to nation-scale problems: Municipal solid waste management in Saudi Arabia“.

Paper Overview (TL;DR)

We applied systems-engineering techniques to investigate how municipal solid waste (MSW) is treated, using Saudi Arabia as a case study. We built a mathematical model of the system and used it to investigate different scenarios.

This model showed that a combination of anaerobic digestion (for food and organic waste), recycling and incineration (for non-recyclable materials) would reduce the total mass of waste to landfill by 73%. Not only that, this approach reduced the effect of the waste management on the environment by 98%.


In Saudi Arabia, municipal solid waste (MSW) is currently disposed of in landfill (dumping it in a hole). This causes irreparable environmental harm and increased hazard from the possibility of landfill fires.

However, the Saudi MSW system is a large, multi-stakeholder (lots of people are interested in the outcome) system. This makes it a great system to develop “nation-scale” (really big) systems engineering tools on.

System Model

My previous work has focussed on modelling rail vehicles and infrastructure, and using these models to investigate how to reduce their emissions. So we took that and applied it to this giant system – the Saudi MSW system:

System overview showing how different factors (government policy, physical constraints and wider system inputs) effect the environmental effects of solid waste disposal.

Then we built a mathematical model of the system, so that we could understand how the narrow system of interest (the waste stream) behaved when we changed some of the inputs. (This involved lots of checking through other people’s work to see how these things should be modelled!) Finally, we came up with a (pretty complex!) system model that looked like this:

Schematic of the detailed system model of Saudi Arabia’s municipal solid waste system.

Then, we made some assumptions about about the makeup of the waste, and the likely population growth over the next 30 years. With that done, we could then answer two basic questions: What happens if we do nothing? What happens if Saudi citizens reduce their waste per person from 1.38kg/day to 1.14kg/day (in line with the UK)? And the answer is:

How much waste to landfill? ‘Baseline’ is if nothing is done. ‘Source reduction’ is if Saudi citizens reduce their municipal solid waste to 1.14kg/day, in line with the UK.

Using the Model

Then, we took this freshly made model and applied it to a range of possible scenarios that we’d devised, based on what we would do, and what other people had suggested:

Scenario numberWhat it involved
Scenario 0 (baseline)Everything to landfill
Scenario 0.1Everything to landfill, but less waste per person (1.14 kg/person/day)
Scenario 0.2Everything to landfill with much less waste per person (1.02 kg/person/day)
Scenario 1Food, glass and metal to landfill. Everything else is incinerated
Scenario 2Everything except glass and metal goes to anaerobic digestion
Scenario 3Anaerobic digestion or recycling for everything. (Incineration for textiles)
Scenario 4Anaerobic digestion or recycling for everything. (Landfill for textiles)
A quick summary of the scenarios used in the study, showing where all of the waste ended up.

And this gave us some pretty graphs, showing how much waste went to landfill in each year for each of the scenarios. (Each of the measures was given 10 years to implement, because making new infrastructure takes time.)

Graph showing how much waste goes to landfill each year under each of the simulation scenarios.

Then, we could look at how much went to landfill in total over the 30-year horizon of the simulation:

The total mass to landfill over the 30-year horizon for each of the scenarios.

This is pretty clear! S3 leads to the lowest total mass to landfill over the 30 years of the study. Job done!

Environmental Impact

Well, almost. Finally, we wanted to check how the environmental effects of each of the scenarios compared. For this we used a commercial life cycle analysis (LCA) tool called SimaPro. We modelled each of the scenarios in this tool, and compared the DALY for each of them. (DALY stands for disability-adjusted life year, or how much of an effect a certain action has on people’s lives. The higher the number, the worse the effect on us.)

The relative DALY (disability-adjusted life year) for each different scenario. These have been normalised so that Scenario 0.1 = 1. For example, in this case Scenario 0 (not doing anything) is 28% worse than Scenario 0.1 (moderate reduction in waste per person per day).

You can see that all of the changes suggested by our systems engineering for waste management study have reduced the hazard the waste poses to people’s lives. In fact, Scenario 3 has only 2% of the impact on people’s lives of the current system. This is encouraging!


The results of this study into systems engineering for waste management are encouraging!

  1. Just reducing the amount of waste people create would reduce the amount of waste to landfill by 89 million tonnes over 30 years.
  2. Scenario 3 (anaerobic digestion of most waste, combined with recycling and limited incineration of non-recyclable materials) is the winner! This would save 328 million tonnes of waste going to landfill over the next 30 years.
  3. Scenario 3 also reduces the impact of waste management on people’s health by 98% compared to the current system.

Other work

For more information on my research, please check out my Research page.


Using Intermittent Electrification to Reduce Rail Emissions

The journal of Railway Engineering Science has published my most recent paper, entitled An investigation into intermittent electrification strategies and an analysis of resulting CO2 emissions using a high-fidelity train model. The paper is open access and freely downloadable.

(Intermittent electrification means electrifying parts of a rail route, rather than the whole thing.)

Paper Overview (TL;DR)

We used a model of a train to investigate how intermittent electrification can help reduce carbon dioxide emissions in the near term. (The long-term goal should be to electrify everything!)

Using intermittent electrification can reduce carbon dioxide emissions by 54%, increasing to 59% if we include the CO2 embodied in the electrification infrastructure.


Electrification is expensive, costing around £1.5 million – £2.5 million per single-track kilometre. It also takes a long time – these are large infrastructure projects that take time to deliver. Instead of continuously electrifying from (e.g.) London outward, intermittent electrification could reduce cost and provide decarbonisation faster.

As we already had a rail vehicle model from previous work, we used this to investigate!

The Rail Vehicle Model

The rail vehicle model was based on the model developed in earlier work. Next, it was extended to include the intermittent electrification to allow it to switch between overhead line equipment (OLE) and the on-board diesel engine.

Intermittent Electrification Model Results

We ran the model over the route from London Paddington to Plymouth in both diesel and electric modes. Next, we compared the two, which showed us where CO2 emissions were the greatest along the route.

A graph showing the carbon dioxide emissions from a train operating between London and Plymouth. Diesel emissions are much higher than emissions from electric running.
The carbon dioxide emissions from different sections of the route between London and Plymouth.

This allowed us to check how electrification would affect the carbon dioxide emissions of the route. We found that 50% of the carbon dioxide savings came from electrification of 30% of the track!

Ranked bar chart showing carbon-dioxide savings from each section of track. 50% of the CO2 savings come from electrification of 30% of the track.
When the sections are ranked by carbon dioxide saving from electrification, 50% of the saving comes from electrifying 30% of the track.

In fact, intermittent electrification produced higher carbon dioxide savings than continuous electrification across the board!

Graph showing how the carbon dioxide saved in intermittent electrification is higher than for continuous electrification at all points.
At 50% electrification, intermittent electrification saves 54% more CO2 than continuous electrification.

Whole-life Carbon Dioxide Emissions

Until this point, we had been looking at ‘operational’ carbon dioxide emissions. (These are emissions that come about from running the trains on the railway.)

However, electrifying track requires steel, concrete and machinery to string up the wires, all of which produce carbon dioxide. So, the next step of our analysis investigated the ‘whole-life’ emissions of the railway, including an estimate for the carbon dioxide emitted during electrification.

Including industry-recognised numbers for the carbon cost of electrification, we found that intermittent electrification saved up to 59% more carbon dioxide than continuous electrification.

Graph showing comparison of the carbon dioxide saved by intermittent electrification and continuous electrification, including carbon dioxide emitted during electrification.
Intermittent electrification can save up to 59% more CO2 than continuous electrification.


This research showed some interesting conclusions, but there are two main caveats to bear in mind:

  1. I am not a life-cycle analysis (LCA) expert. I have taken industry-standard numbers and used them in this analysis.
  2. There are operational and logistical difficulties with electrifying intermittently. Not least, electrifying in this way may require more transformers/electrification infrastructure than doing so in a continuous manner.

With these in mind though, this paper goes some way toward making the case for intermittent electrification to speed up the decarbonisation of the UK’s rail network.

Other Work

For more information on my research, please check out my Research page. If you’re interested particularly in rail or road vehicles, then please check out the relevant pages.


Digital Displacement for Hybrid Trucks

The Proceedings of the IMechE, Part D: Journal of Automobile Engineering has published a new article of mine, entitled Modelling and Experimental Validation of the Performance of a Digital Displacement® Hydraulic Hybrid Truck. I did this work in partnership with Danfoss. The paper is open-access and freely downloadable.

Paper Overview (TL;DR)

We added hydraulic hybridisation to a light truck and improved its fuel economy by 23%. We also used simulation to show that hydraulic hybridisation can improve the fuel economy by up to 95%, depending on the type of driving


My previous work has shown that hydraulic hybridisation is effective at reducing the fuel consumption of heavy goods vehicles. However, this research was limited due to the hardware used and wasn’t tested in a controlled environment (on a rolling road, or ‘dynamometer‘).

In this work, Danfoss converted a light truck to a hydraulic hybrid using their Digital Displacement® technology, and tested it on a dynamometer.

The Vehicle Conversion

Danfoss converted a light truck to a hydraulic hybrid using a Digital Displacement® pump/motor. (A pump/motor can produce negative [braking] torque and a positive [accelerating] torque, rather than just one or the other.) The system stores recovered energy in a high-pressure hydraulic accumulator.

A photo of the underside of the converted hybrid truck, showing the power take-off, connecting shaft, DDPM, gearbox, driveshaft and differential.
The underside of the converted vehicle.

Test Results

First, we checked to see how consistent the drivers were. They were remarkably consistent; the distance covered in the tests had a standard deviation of 0.3%! After that, we checked the results from the logging equipment to make sure it was sensible.

Results from the hydraulic hybrid truck model showing the vehicle speed and pressure in the high-pressure accumulator.
Test results showing how the pressure in the high-pressure accumulator varies with speed. Looking good!

Next, we looked at how the fuel usage of the hybrid vehicle compared to the un-altered vehicle. (Hint: it used a lot less fuel!)

Timeseries of fuel usage for the diesel truck versus the hybrid truck. The hybrid truck uses 19% less fuel.
In testing, the hydraulic hybrid truck used 19% less fuel!

Model Validation

After that, we checked to make sure our model was a reasonable fit with the experimental data .(This step is called ‘validation‘, and is a common technique in engineering.)

Graphic comparing speed, engine speed and total fuel for the measured and simulated diesel truck, showing that the traces match.
Making sure the model and the test data match. Looking good!

Model Performance

Now that we had a working model, the next step was to use this model to investigate what benefits might be achievable if we could change some of the system parameters. We found that with some small parameter changes, and by turning off the engine when it wasn’t being used, the vehicle would use much less fuel. It used 36% less fuel over this cycle!

Results from our model of the hydraulic hybrid truck. The results include speed, pressure in the high-pressure accumulator and the total fuel used.
Our modelling showed that the hybrid truck used 36% less fuel over the JE05 driving cycle!

Next, we looked at the performance on a more common drive cycle, the WLTP cycle. In this case the hybrid truck used 19% less fuel than the original diesel vehicle – not bad!

Results from running the hydraulic hybrid truck model over the WLTP drive cycle. The included values are speed, pressure in the high-pressure accumulator and total fuel used.
Modelling results showing that the hybrid truck uses 19% less fuel over the WLTP drive cycle.


The main conclusions of our paper were:

  1. It’s possible to use Digital Displacement technology to make a hydraulic hybrid truck.
  2. Hybridising the vehicle reduces fuel consumption by 19-36%, depending on the drive cycle.

If you want more information, please check out the full paper online.

There is more information on my other decarbonisation work on electric vehicles and rail vehicles as well.


High-Fidelity Bi-Mode Rail Vehicle Model

The Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit has published a new article of mine, entitled Development and control of a rail vehicle model to reduce energy consumption and carbon dioxide emissions. The paper is open-access and freely downloadable.

Paper Overview

We made a detailed model of a bi-mode rail vehicle and used control to reduce its carbon emissions (decarbonisation).

The Rail Vehicle Model

A bi-mode train is one that can run on electricity from overhead wires (sometimes called overhead line equipment or OLE), or by generating electricity from the on-board diesel generator. Then we used vehicle data to make sure the model represented the behaviour of the real train.

Control for Decarbonisation

We used that model to look at a range of controllers for reducing the CO2 emitted by the train (decarbonisation) during its run from London Paddington to Plymouth in the UK. Our controller reduced the CO2 emissions of the train by 19%. Without this controller, the train produces around 42g of CO2 per passenger-kilometre of diesel running.

Intermittent Electrification

We also investigated some simple ‘intermittent electrification’, where only some sections the route are electrified. (Some people also call this discontinuous or discrete electrification.) You can see the results of this analysis in the picture, below.

A graph showing the decrease in emissions possible if intermittent electrification is used with bi-mode rail vehicles.
This figure shows the key conclusions of the paper. Firstly, that electrification will reduce the CO2 emissions of the route (but not to zero, because the UK still produces electricity using some fossil fuels). Secondly, that intermittent electrification (only electrifying some pieces of track) and using selective engine shutdown can reduce the CO2 emissions when only part of the track is electrified.

The figure shows that intermittent electrification can be more effective than continuous electrification at reducing carbon dioxide emissions. It also shows that selective engine shutdown can be used for the decarbonisation of a bi-mode rail vehicle.

If you’d like more information on my work on railway decarbonisation, check out the project webpage, or get in touch!


This project was generously funded by the RSSB (project COF-IPS-02). We would also like to thank Hitachi RailAngel TrainsGreat Western Railway and Chiltern Railways for their support.