ML can potentially reduce industrial emissions by helping to streamline supply chains, invent cleaner materials and chemicals, design for fewer materials, improve production quality, predict machine breakdowns, optimize heating and cooling systems, and prioritize the use of clean electricity over fossil fuels.
ML can have a substantial positive climate impact on the industrial sector under the following circumstances:
- When there is enough accessible, high-quality data around specific processes or transport routes.
- When firms have an incentive to share their proprietary data and/or algorithms with researchers and other firms.
- When aspects of production or shipping can be readily fine-tuned or adjusted, and there are clear objective functions.
- When firms’ incentives align with reducing emissions (for example, through efficiency gains, regulatory compliance, or high GHG prices).
Yet greater efficiency may increase the production of goods and thus GHG emissions (via the Jevons paradox) unless industrial actors have sufficient incentives to reduce overall emissions.
Machine Learning Application Areas
Optimizing supply chains
- Freight Consolidation: Optimizing the shipping and transportation of goods across today's globalized supply chains is a complex, multifaceted challenge – especially for perishable goods. Bundling shipments together through freight consolidation can dramatically reduce the number of trips and associated GHG emissions. ML can optimize complex relationship between the various dimensions involved in shipping decisions, such as shipment mode and origin-destination pairs.
- Goods Demand Forecasting: The production, shipment, and climate-controlled warehousing of excess products is a major source of industrial GHG emissions. ML may be able to mitigate overproduction and/or the overstocking of goods by improving models for forecasting consumer demand, especially for perishable goods or "fashionable" items that quickly become obsolete.
- Greenhouse gas emissions mapping for supply chains: While some factories release publicly-available data on their emissions, this data is not available (or is misreported) in many cases, especially in emerging markets. ML can help map greenhouse gas emissions using a combination of remote sensing and on-the-ground data, leading both consumers and upstream suppliers to make greener decisions in sourcing products.
Developing cleaner products and materials
- Accelerated science for clean energy technologies: Designing new materials is important for many applications, including energy storage via batteries or solar/chemical fuels. ML can help suggest promising materials to try, thereby speeding up the materials discovery process.
- Generative Design: ML-enabled 3D-modelling software can can help create new designs for physical structures that reduce the need for carbon-intensive materials – especially cement and steel.
- Additive Manufacturing: Novel manufacturing techniques such as 3D printing allow for the production of unusual shapes that use less material or different types of material, but may be impossible to produce through traditional manufacturing methods.
Optimizing factory operations
- Supply and demand forecasting: The supply and demand of power must both be forecast ahead of time to inform electricity planning and scheduling. ML can help make these forecasts more accurate, improve temporal and spatial resolution, and quantify uncertainty.
- Methane Leak Detection: Fertilizer factories and some other chemical plants often leak methane, a powerful greenhouse gas. ML can help detect and prevent these leaks.
- Adaptive Systems Control: ....
- Predictive Maintenance: Quickly detecting machinery faults can help reduce electricity and materials waste. ML can help detect faults in real time from machinery sensor data, or even forecast them ahead of time to enable preemptive repair or replacement.
- Demand response: by reducing or shifting their electricity usage during peak periods in response to time-based rates or other forms of financial incentives.
- Rebitzer, G. et al. Intro to life-cycle analysis (2004)
- Gao, J. et al. Google’s white paper on ML for data center optimization
- Gustavsson, J. et al. Intro to food waste (2011)
- Wikipedia - Intro to Industry 4.0
- Carbon180 - Building a New Carbon Economy (substantial overlap with carbon sequestration and agriculture)
Online Courses and Course Materials
Autodesk Sustainability Workshop: "free online learning resources that teach the principles and practice of sustainability in engineering and design. This site was created to help students, educators and professionals learn more about sustainability in engineering and architecture professions." Practical videos on sustainable product design, materials selection, and engineering available here.
Conferences, Journals, and Professional Organizations
Major professional organizations
IEEE Power & Energy Society: "[T]he world's largest forum for sharing the latest in technological developments in the electric power industry, for developing standards that guide the development and construction of equipment and systems, and for educating members of the industry and the general public." Website here.
- Ellen MacArthur Foundation -- consortium around the circular economy and sustainable supply chains/materials development
- Helps (mostly large) manufacturers consider the entire life cycle of their products Website here.
- The BlueGreen Alliance's Clean Economy Manufacturing Center (USA) helps American manufacturers and construction firms improve efficiency and sustainability. Website here.
Libraries and Tools
- The Materials Project
- Inorganic Crystal Structure Database
- SciFinder (paid)
- USA Industrial Assessment Centers: database of solutions from university collaborations with SMEs to save energy and reduce costs
See the subpages listed above for application-specific datasets.
Relevant Groups and Organizations
Sustainability Accounting Standards Board (SASB) -- popular global standards board for corporate sustainability efforts Website here.
- Carbon180 New Carbon Economy Consortium is focused on carbon capture and sequestration, including industry, agriculture and forestry. Website here.
- Davis, Steven J.; Lewis, Nathan S.; Shaner, Matthew; Aggarwal, Sonia; Arent, Doug; Azevedo, Inês L.; Benson, Sally M.; Bradley, Thomas; Brouwer, Jack; Chiang, Yet-Ming; Clack, Christopher T. M. (2018-06-29). "Net-zero emissions energy systems". Science. 360 (6396): eaas9793. doi:10.1126/science.aas9793. ISSN 0036-8075.
- Mike Gualtieri, Noel Yuhanna, Holger Kisker, Rowan Curran, Brandon Purcell, Sophia Christakis, Shreyas Warrier, and Matthew Izzi. The Forrester Wave: Big data streaming analytics, Q1 2016. Forrester.com, January 2016.
- Kazi, Rubaiat Habib; Grossman, Tovi; Cheong, Hyunmin; Hashemi, Ali; Fitzmaurice, George (2017-10-20). "DreamSketch: Early stage 3D design explorations with sketching and generative design". Proceedings of the 30th Annual ACM Symposium on User Interface Software and Technology. New York, NY, USA: ACM. doi:10.1145/3126594.3126662. ISBN 978-1-4503-4981-9.
- Richard Evans and Jim Gao. DeepMind AI reduces Google data centre cooling bill by 40%. DeepMind blog, 20, 2016.
- Zhang, Xiao; Hug, Gabriela; Kolter, J. Zico; Harjunkoski, Iiro (2016-07). "Model predictive control of industrial loads and energy storage for demand response". 2016 IEEE Power and Energy Society General Meeting (PESGM). IEEE. doi:10.1109/pesgm.2016.7741228. ISBN 978-1-5090-4168-8. Check date values in:
- Berral, Josep Ll.; Goiri, Íñigo; Nou, Ramón; Julià, Ferran; Guitart, Jordi; Gavaldà, Ricard; Torres, Jordi (2010). "Towards energy-aware scheduling in data centers using machine learning". Proceedings of the 1st International Conference on Energy-Efficient Computing and Networking - e-Energy '10. New York, New York, USA: ACM Press. doi:10.1145/1791314.1791349. ISBN 978-1-4503-0042-1.
- Sorrell, Steve (2009-04). "Jevons' Paradox revisited: The evidence for backfire from improved energy efficiency". Energy Policy. 37 (4): 1456–1469. doi:10.1016/j.enpol.2008.12.003. ISSN 0301-4215. Check date values in: