On the Future of Sustainable Design


This paper is a general roundup of underused elements for sustainable building design. The results of several simulations revealed strategies for designing energy-efficient, financially viable, and aesthetically acceptable buildings. The less conventional ideas include radiation absorption through the colour of roofs and walls, minimization of the surface-area-to-volume and ground-area-to-volume ratios, and the adjustment of building orientation. For ideal performance, these strategies should be used in combination to maximize thermal absorption in colder months as well as thermal emission in warmer months. These methods include both passive (through the use of colour, shape, and insulation), and active ones (through automatically adjusting heating systems).


It is a curious and hitherto unprecedented fact of geology that we have entered a new era of our own making: the Anthropocene1. The 1987 report of the World Commission on Environment and Development named Our Common Future, which was perhaps the first official recognition of a need to change in accordance as a species, states: “The Environment does not exist as a sphere separate from human ambitions, and needs.”2 Due to low recognition of this, this environment is now changing, and the need arises to adapt our architectural design in accordance to this new age. In a school project the author participated in, and upon which this paper is based on, the task was to design a school building with several criteria focused on sustainability.

Throughout this investigation, sustainability shall be defined as the approach that minimizes cost and energy.


Energy efficiency is improving in today’s technology and will continue doing so, gradually reaching towards a limit determined by the laws of physics. Utilizing how heating and cooling follow the same principle, the Tesla Model Y is produced and sold with a heat pump for its battery to exchange charge more efficiently3: yet another example of technology marginally improving to approach ideal performance. Therefore, engineers must resort to the laws of physics to determine a design’s limits.

There are two options when optimizing a system. Since perfection is an impossibility, a direction of error must be chosen. In the case of optimizing for expending energy on maintaining a temperature, this can mean either spending more energy on heating or cooling. Given thermodynamics, the former seems the more efficient, as heat is usually a fundamental byproduct of cooling, especially by electrical means. Hence, the designs below are aimed to not require any cooling.

Furthermore, while it’s by no means a novel idea (many Norwegian buildings employ the tactic), shaping the building as circular, or with as many obtuse angles as possible, may have a surprising effect on the buildings’ efficiency.

Finally, the system is limited by current technology. This is irrelevant to the design though, as it is a simulation, and thus should be looked at comparatively to itself rather than any real-world counterparts. All values used for variables such as a window’s R-values and solar panel efficiency used in the simulation are hence made irrelevant.


To test out the hypotheses of surface-area-to-volume ratio, colour, and others, the author used Energy3D, a simulation software created by Charles Xie, Corey Schimpf, and Jie Chao et al. A few peremptory notes of caution are called upon: first and foremost, while an incredible learning tool, it is meant merely “as an entry-level energy simulation tool” 4 rather than an accurate one. This is most clear in the energy simulation itself: depending on the day the simulation begins, the total energy spent in a year may vary. While the source code is not available for examination, this is likely a computational shortcut to make the somewhat slow software run faster: an extrapolation of twelve days, one every month, rather than a calculation of the energy consumption of every single day of a year. To keep this from interfering with the results, the details of the stimulation have been standardized for the various designs. This does mean though, that all the upcoming numbers are not meant as accurate representations: rather they should be considered comparable to one another.

The goals of the design of a small school, as specified in the task, included:

  • a limit to the cost of 3 500 000 NOK, roughly equivalent to 300 000 £
  • a lot size limit of 900 m2
  • a list of required rooms, all of which are included in Appendix II
  • general building requirements including fire and safety regulations, such as a minimum of two entrances to each room and a minimum roof pitch angle
  • the latitude was set to that of Oslo
  • there were also limits that only one solar panel was allowed to encourage true sustainable design, rather than increase in roof area and cost for solar panels
  • room thermostats should be set to at least room temperatures (21°C for consistency) for the time the rooms are occupied
  • there was a maximum of 20 000 for the entire structure to motivate sustainable design

The design was first drawn on paper to meet all the geometrical tasks. Out of three designs, one design had only one floor and turned out too large, while another was impossible to model in the software (due to difficulties with no interior walls and no second floors), so the third design was settled on as a compromise. The design can be found in appendices II and III, though as mentioned prior, these are not of any importance as it was kept constant for all the simulations. Then, comparisons were made with the same model but with various elements missing to see if improvements were made.


Most importantly for the project, the design fits all the criteria. It consumed 14 294.4 0.1 kWh, and cost 257 325 £, given the software’s original settings. The monthly energy consumption can be found in Appendix I.

Table 1: Simulated yearly energy requirements in

Not Insulated

Night Heating on

White walls, red roof

Solar panel roofing

No windows

South – facing


Final design

64 732

27 107

25 349

-2 064

13 040

16 117

18 689

14 294

The following is a discussion of the specific values:

As the numbers are all given by the software to the nearest whole, they should have an uncertainty of 1 kWh

The biggest change to the energy consumption was (somewhat predictably) an automatic heating system. The temperature during working hours was kept at 20 . During the night (to have the most room for error, counted in the period 22.00 to 7.00), heating was taken to the lowest temperature possible without starting the cooling system, as the software did not have an option to turn the heating systems off. Removing this feature of the building increased the annual net energy to 27 107 kWh, almost doubling the initial energy. This was the most productive change in the design.

Changing the building colour to a more conventional one (completely white exterior walls and red roof) had a much higher energy consumption of 25 349 kWh. This is mostly because, in the summer, the cooling system had to consume a lot more energy, while the original design had no cooling costs at all. This finding supports the idea that designing a building to radiate heat without a cooling system is more efficient than designing a building to equalize cooling and heating requirements. However, the simulation does not rule out the possibility to optimize both heating and cooling to further decrease energy requirements. A rendering of the winter-oriented colouring may be seen in Appendix III.

The solar panel was responsible for generating 2 064 kWh. The solar panel efficiency, given by the software in percentages, was increased to 24%. Given current photoelectric technology, this is a very optimistic development, as the industrial average was only 20% in 20175. With the entire Southern-facing roof covered in solar panels, they generate more energy, though not quite enough to make the building net positive.

Perhaps counterintuitively, removing all windows (and allowing the cooling system to use 1 445.7 kWh of energy through the year) reduced the energy to 13 039.6 kWh. However, this building was optimized for the presence of as many windows as possible without a cooling system.

Partially to test the software, removing the inside walls took the energy level needed down to 8 807.6 kWh, which could be an indication that larger rooms are more energy-efficient. It is, however, a very questionable result, as the software may be understanding each of the rooms as individual houses that need to be heated in addition to the outer house. The inside walls are a constant in the other tests, so the design with the rooms was kept constant for all the variations of the simulation.

Optimizing the yaw angle saved 1 823 kWh. While the initial idea was that the longer side of the house, along with most windows, doors, and solar panels should be facing south (or whichever direction the sun is at its highest), it turned out that the ideal angle is approximately in the south-south-west. This is likely due to the building’s irregular shape, an elongated octagon. This orientation, found by iterative optimization, is likely the balance of the building absorbing the most sunlight, including mornings and evenings. The perfectly south-facing orientation required an annual net energy of 16 117.2 kWh.

An approximation of the same design (with the same volume) without the octagonal shape used 18 688.7 kWh of energy. This increase was mostly due to the cooling system in the summer, though this system only used 3 944.3 kWh of energy, and even just subtracting this from the total (which is not a fair comparison), the energy used was 14 744.3 kWh, about five hundred kilowatt-hours more than the octagonal shape. A regular octagon, or ideally, a circular building should be even more efficient, though this was not simulated due to the need for extensive rework of the design, which would not be consistent with the other tests, along with limitations of the software.

Perhaps most obviously, insulation is important. Changing fiberglass and triple pane glass for cellulose and single-pane glass takes the annual energy consumption all the way up to 64 732 kWh. It takes the cost down to 205 545 £, and the breakeven point of insulating everything is only thirteen years (using the electricity price value provided above). The windows are the largest source of the increase, changing just the windows to triple-pane glass takes the energy consumption down to 21 624.8 kWh.


The idea behind roof colour was the following. In warmer months, this in the northern hemisphere being the summer months, the sun rises much higher in the sky than in colder months. Therefore, the roof, when exposed to more sunlight should be more reflective than the walls, while the walls when less exposed, should absorb more. Colouring them accordingly is a passive way of thermal regulation that negates the need for more active methods that expend energy. It’s by no means a novel idea, despite which light roofs still seem unfashionable. That being said, fashion changes with exposure, and exposure changes with functionality. The final design, as seen in Appendix III, takes the idea to the extreme. In some parts of the world, such as the state of California there are already building codes for roof colour7, a strategy proven effective by these simulations.

The same applies to other changes. Though the author is not familiar with any regulations mandating circular buildings, the idea behind maximizing the surface-area-to-volume ratio will minimize the building’s energy losses. Perhaps future buildings in ever-increasing imperfect environments (such as warm or cold deserts) are bound to be hemispherical.

Concerning the solar panels on the building, as seen in Appendix I, the majority of this energy would be generated in the summer. As the cooling system has been made intentionally obsolete, this energy would be fed back to the local grid. With today’s battery technology, this is not very practical. It should also be noted that while the smaller panel cost 3 900 £, and the larger cost 35 000£ (these prices are rounded to two significant figures). Assuming the cost per kWh to be 0.994 £ 6, the difference in the initial cost of 31 100£ would take more than thirty years to break even. Despite this, the building is without need for external power from March to October, and after the breakeven point, the building would cut running costs by more than a thousand pounds sterling a year.


There are a few conclusions that can be drawn from the trends in the data. The first one is that the greatest single thing any household should do to save energy and money is to insulate. Windows are the largest heatsink, though this conclusion may be biased by this design’s prominent use of windows. Furthermore, for public buildings not visited during night time, shutting off or turning down the heating system is a major part of saving energy.

The hypothesis regarding the higher efficiency of neglecting the need for cooling was confirmed as well. Though there is certainly a cheaper balance, the general conclusion is that if the building needs to cool down, it generally takes more energy than if it needs to warm.

Creativity should be used in the colouration and shape of buildings: rounder, or if that’s too complicated in construction, octagonal shapes in buildings are both more energy-efficient and allow for a more unconventional and interesting design. Windows should be avoided, though if designed with care, windows can also save energy by making cooling systems obsolete. And finally, simply orienting the building can have a surprising effect on warming costs.

There are many different types of products out there, but few can compare to the usefulness and durability of durable materials. This is why so many industries have turned to these materials when it comes to the manufacturing process. Read all about the most durable materials on The reason that most people like to use these materials is because of how long they can last and the fact that they don’t cause any sort of pollution. These are a few things that you should be taking a look at when it comes to getting a durable materials review so you can get an idea of what all good products for sale are made out of.

Extension for research

What other possibilities for sustainable design are available? There are a few that should be discussed. The following are possibilities that the simulation could not verify:

The software appeared to ignore space within the roof. It is conceivable that this could effectively be an entire third floor, had the software simulated it. Otherwise, the roof could be lowered significantly, as in the center the building is 7.2 meters tall (due to pitch requirements), while the two floors are only 3.2 meters each. An additional floor would be a 50% increase in a usable area without any increase in building footprint, a worthwhile investment.

Furthermore, another conceivable solution is to go down rather than up. There’s a very large space below street level, perhaps even, when the technology to build large structures beneath the surface is developed, easier accessible for construction of complex structures than upwards. The ground is also a great insulator, though of a lower than the ideal temperature that would have to be insulated. Despite these seeming benefits, not many buildings develop subterranean sections beyond perhaps a basement or an underground parking lot.

Overall, there are many possibilities for more sustainable designs. Some may currently look uncomfortable or ugly, but opinions change with exposure. Perhaps the future holds black walls with white roofs. Nonetheless, the current architecture will almost certainly change, for better or worse. Much like we’re moving into a new geological age, we may see a new architectural movement: hopefully a more sustainable, creative, and engaging one.


First of all, the software “Energy 3D” was created by Charles Xie, Corey Scimph, Jie Chao, Saeid Nourian, and Joyce Massicotte, and it was central for this project. The author would like to thank the two schools of Willibald-Gluck-Gymnasium for participating in the project and furthering the design with scaled models and automation. The author would also like to thank Lucas Kuhn, the partner in the design of the project, and Gerard Decelles, our teacher of science, for his priceless help and guidance in the project.


1 – Zalasiewicz, Jan, Mark Williams, Will Stefen, and Will Crutzen Paul Stefen. “The New World of the Anthropocene.” Environmental Science & Technology (ACS Publications) 44, no. 7 (February 2010): 2228 – 2231.

2 – World Commission on Environment and Development. “Our Common Future”. Report of the UN WCED, Oslo: Oxford University Press, 1987.

3 – Lang, Fabienne. “Tesla Model Y Heat Pump, ‘Best Engineering’ Elon Musk’s Seen in a While”. 24 March 2020. (accessed August 7, 2020).

4 – Xie, Charles; Schimpf, Corey; Chao, Jie; Nourian, Saeid, and Massicotte, Joyce. “Learning and Teaching Engineering Design through Modeling and Simulation on a CAD Platform.” Computer Applications in Engineering Education, 26(4), June 2018: 824 – 840.

5 – VonderHaar, Grant. “Efficiency of Solar Cell Design and Materials”. Missouri S&T’s Peer to Peer, (2). May 2017.

6 – Statisk sentralbyrå. “Electricity prices”. Governmental statistics bureau report, Oslo: Statistics Norway, 2019.

7 – Roosevelt, Margot “New anti-warming tool: White Roofs” 10 September 2008. (accessed August 17, 2020)

Appendix I: The monthly energy expenditure of the final design

Appendix II: The design of the final design

1m __

Each floor was 3.2m tall.

A few notes: this design was driven more by the criteria, the aesthetics, and most of all: the limitations of the software, much more so than the physics. For instance, there was no logical way of simulating a second floor with a layout different to the first floor. Therefore, this is included more for replicability than showing any point. Each floor was 3m tall, and the roof was 4m tall. Perhaps one important note was that making the inside volume makes the whole system much more efficient, and perhaps there will be more roof-integrated housing with lowered upper floors in the future to account for this. The author’s group was not allowed to include this in their design due to the criteria and the possibility that the software doesn’t count roof volume as internal volume.

Appendix III: The software’s rendering of the final design

See note in Appendix II.

About the Author

Alexander Mitrokhin is currently a High School student at Skagerak International School in Norway, applying for university in 2021. He is interested in the future of science and engineering, with a focus on spaceflight and astrophysics. Currently also interested in fiction, both science fiction and fantasy, both for entertainment and exploration of ideas.

Leave a Reply

Your email address will not be published. Required fields are marked *