How to use artificial intelligence to achieve environmental sustainability

Why is essential?

Proactively managing climate action and environmental topics is a central part of Ericsson’s sustainability strategy. is one of three main focus areas for sustainability – the other two being responsible business and digital inclusion (read more, here). With the growing threat of global warming, the negative impact of carbon emissions is an urgent global concern. The pressure on businesses to accelerate climate action and limit global warming has never been greater and the corporate world is committed to realizing its ambition to become Net Zero across its entire value chain. . According to U.S. Environmental Protection Agency Climate Change Report 2019transport accounts for an alarming 29% of global greenhouse gas (GHG) emissions. At Ericsson, we take the need to decarbonize seriously. To meet the threat head on, Ericsson is committed to achieving net zero emissions across our value chain by 2040. Ericsson is already working towards a major first step to reduce emissions by 50% across the value chain. supply and portfolio by 2030 and become Net Zero in our own operations at the same time. One of the main activities to reduce supply chain emissions is the transportation of products.

Instead of the above, and to achieve our ambitious goal of reducing CO2 equivalent emissions, experts from the IT AI & Automation group, together with Supply, have defined a to optimize the transport of products – including monitoring, prediction and reduction – with the ambition to apply AI to make the unimaginable possible.

How to measure and analyze CO2 equivalent emissions from transport

But how could this strategic plan be implemented to actually make a difference in transportation emissions and areas of business outside a company’s direct control? First, we needed information to understand the full scope of supply chain emissions through measurable data and transparent reporting, called the “Monitoring” phase.

This phase mapped the current CO2 equivalent emissions scenario in the organization, including multiple complex product transportation streams such as customer supply chain, product supply chain, local transport and various processes. The main challenge in developing such a solution was the lack of data availability, related data from various sources and developing the precise logic to calculate the CO2 equivalent emission. Using various analytical techniques and methodologies based on fuel, distance and cost, we were able to calculate the emissions associated with transportation. After much trial and error, the distance-based method turned out to be the best-suited approach for Ericsson Transport Management. We derived the CO2 equivalent emission by modeling frequent parameters such as the volume of goods purchased, the distance travelled, the standard emission factor for the respective mode and/or type of transport, etc. The model has been built generic enough to fit most similar transport services. A simplified version of the calculations of CO2 equivalents for different modes of transport is as follows:

Y=Σ (mass of goods purchased (tons or volume) × distance traveled in the transport segment (km) × emission factor of transport mode or type of vehicle (kg CO2e/ton or volume/km))

With the CO2e emission algorithm and LowCode web-based visualization dashboard allowing global users to interact at the same time, we were able to deploy a complete monitoring solution based on the data model, a snapshot dashboard reference as below:

CO2e Emissions Dashboard - Ericsson Global and Units sage.  Note: All figures shown are dummy data.

Figure 1: CO2e Emissions Dashboard – Ericsson Global and Units sage. Note: All figures shown are dummy data.

Ericsson Global CO2e Emissions Dashboard - Transport mode wise.

Figure 2: Ericsson Global CO2e Emissions Dashboard – Transport Mode Wise.

In addition to measuring and monitoring CO2e emissions in different modes of transport, we were able to improve data quality. By analyzing the data collected, the company was able to identify specific areas where data quality was low and could lead initiatives with data stewards to improve the quality of data collected and initiatives at the operational level to capture the right data. . This iterative process of improving data quality will gradually help business drivers make sound decisions.

Transforming data into future information: prediction phase

With the basic data and analysis in hand, the next logical step was to capture important patterns and trends to predict future business, noted as the “prediction” phase.

Forecasting volumes and weight shipped for various transportation routes was a complex process due to the high level of disaggregated commodity flows. The uncertainty of transmission services, the diversity of processes and non-standard procedures made it difficult to optimize the use of transmission resources and distribution planning.

With the application of machine learning (ML) techniques such as regression, clustering, deep learning, etc., and using historical and transactional data, we have developed a long-term shipment weight forecast. and in the short term more accurate than that which we obtained with manual predictions.

Applying such approaches not only reduces the need for manually generated forecasts, but also helps Logistics Service Providers (LSPs) to have better delivery accuracy, leading to improved rates and hence lower costs. The forecast will ensure the availability of transport capacity and significantly reduce the delivery time.

By using a good forecast, LSPs can identify key drivers of shipping freight across regions and shipping lanes, as well as the impact on the overall product transportation chain.

The modeling part of the solution is composed of several boosting algorithms having a wide range of hyperparameter tweaks on features like learning rate, max_depth, n_estimators, subsample. Due to volatility and inconsistencies in the data, no single model could produce results, so a set of machine learning models was developed with different hyperparameters. The framework was designed in such a way that the best models (for the tech-savvy, this was implemented via lowest WMAPE, weighted average error in absolute percent) would be dynamically recovered during l execution and would be used to predict the associated weight/volume.

Prediction to actions – Reduction phase

After measuring and analyzing the results of CO2e emissions and having good long and short term forecasts, it is now time to prepare plans and implement methods to reduce CO2e emissions, also called the ” reduction”.

With the help of monitoring and forecasting, continuous warning processes can certainly contribute to the iterative optimization of CO2e emissions.

  • Reduce and improve transport activities
    • Fleet optimization – Higher degree of filling in trucks, reduction of unnecessary air shipments and updating of the fleet.
    • Better planning of packaging and transport material, better collaboration with suppliers.
  • Improving transport efficiency
    • Avoid short lead times by using forecasts
    • Use well-organized navigation
    • Predictive Analysis to Prevent Vehicle Breakdowns and Efficiently Use Less Energy (Predictive Maintenance)

In briefThe transport logistics sector is one of the main consumers of fossil fuels, and is therefore a major contributor to total greenhouse gas (GHG) emissions, representing total GHG emissions, according to the U.S. Environmental Protection Agency Climate Change Report 2019. From our own operations, we have learned that AI can be used to reduce the use of transport vehicles by optimizing vehicle flow, providing more efficient navigation and facilitating shared transport.

Based on our knowledge of the potential of AI and the continued development of industry automation, this blog post has highlighted our three-phase strategic approach – monitoring, predicting and reducing.

In addition to improving the visibility and transparency of CO2e emissions, as well as improving data quality, operational efficiency and customer satisfaction, transport volume forecasting aims to reduce both CO2e and operational expenses.

We believe that the challenges of environmental sustainability could be met through this three-phase approach, coupled with strong domain knowledge. If used wisely, we have no doubt that AI will accelerate our sustainability efforts.

AI is already becoming the key to empowering governments, organizations and individuals to make more conscious decisions and work towards creating a healthier planet. At Ericsson, we work for this cause and are proud to be able to demonstrate how applied AI can bring about lasting change.

The severity of climate change as well as the potential of artificial intelligence makes it too essential not to try, don’t you think?

Learn more

For more AI/ML use cases, please check Artificial Intelligence/Machine Learning (Internal link)


Leave a Reply

Your email address will not be published.