• International Journal of Technology (IJTech)
  • Vol 12, No 5 (2021)

Ship Energy Efficiency Management Plan Development Using Machine Learning: Case Study of CO2 Emissions of Ship Activities at Container Port

Ship Energy Efficiency Management Plan Development Using Machine Learning: Case Study of CO2 Emissions of Ship Activities at Container Port

Title: Ship Energy Efficiency Management Plan Development Using Machine Learning: Case Study of CO2 Emissions of Ship Activities at Container Port
Ivan Dewanda Dawangi, Muhammad Arif Budiyanto

Corresponding email:


Cite this article as:
Dawangi, I.D., Budiyanto, M.A., 2021. Ship Energy Efficiency Management Plan Development Using Machine Learning: Case Study of CO2 Emissions of Ship Activities at Container Port. International Journal of Technology. Volume 12(5), pp. 1048-1057

39
Downloads
Ivan Dewanda Dawangi Department of Mechanical Engineering, Faculty of Engineering, Universitas Indonesia, Kampus UI Depok, Depok 16424, Indonesia
Muhammad Arif Budiyanto Department of Mechanical Engineering, Faculty of Engineering, Universitas Indonesia, Kampus UI Depok, Depok 16424, Indonesia
Email to Corresponding Author

Abstract
Ship Energy Efficiency Management Plan Development Using Machine Learning: Case Study of CO2 Emissions of Ship Activities at Container Port

Ship energy management has an effect both on cost efficiency and the environment due to the huge amount of CO2 emission caused by ship activities. Meanwhile, research regarding efforts to save energy consumption in the container terminal area is scarce. This paper aims to estimate CO2 emissions from ship activities in the container port. The influential variable of CO2 emissions is in consideration to Ship Energy Efficiency Management Plan (SEEMP). The estimation of C02 emission starts from the ship activities when the ship approaches the port, which includes ship maneuvering and ship berthing.  Ship’s energy consumption and CO2 emission were analyzed using random forest regression (RF) at the default setting, and then the effectiveness was verified using k-folds cross-validation. The analysis result showed there are five influential variables to reduce the CO2 emission: (1) main engine power; (2) auxiliary engine power; (3) waiting time in a port basin; (4) maneuvering time; and (5) berthing time. Among those five variables, maneuvering, waiting in a port basin, and berthing have the same position at the top with the same amount of weight importance from the four attribute selection training results. The random forest model training and k-folds cross-validation confirmed that the model has 98.85% of accuracy. Finally, a fuel-efficient operation is discussed, and it can be concluded that by combining several voyage optimizations with a skilled operator and cold ironing when available, it is possible to reduce the CO2 emission by 20%. The findings and proposed plan in this paper can become a reference to develop Ship Energy Efficiency Management Plan.

CO2 emission; Machine learning; Random forest regression; SEEMP; Ship maneuvering

Introduction

    Ship energy management has an effect not only on cost efficiency but also on the environment due to the huge amount of CO2 emission caused by ship activities. Reducing energy consumption has direct impacts on emissions, minimizes the environmental effect, and reduces operational costs. CO2 emissions require reduction to improve port air quality, making emissions factor descriptions necessary (Budiyanto et al., 2019). Unfortunately, research regarding efforts to save energy consumption in the container terminal area is still rarely conducted (Budiyanto and Shinoda, 2020). While it requires development in energy management and operational planning in real time, most research does not discuss the relationship between working time, idle time, and energy consumption of each piece of equipment in depth (Iris and Lam, 2019). The study about energy consumption in the container port was carried using using modality movement which is provide a consistent estimation for carbon emission (Huzaifi et al., 2020). Another aspect that contributed to the CO2 emission is the layout of the container terminal tends to have different results between perpendicular and parallel layouts (Budiyanto et al., 2021).

Ship Energy Efficiency Management Plan (SEEMP) is a ship energy usage plan used as a milestone for energy efficiency development by ship owners and should reflect efforts to improve the ship’s energy efficiency through four steps: planning, implementation, monitoring, evaluation, and improvement. Many companies have already applied environmental management systems under ISO 14001, which contain procedures for selecting the best measure for a particular vessel and then set goals for parameter measurement, control features, and relevant feedback (Witten et al., 2011). Noon data reports, which provide information on ship’s fuel consumption, speed, and weather condition, are the traditional estimation method used for SEEMP development. Unfortunately, it is not fully utilized by shipping companies and is collected only for regulation due to limitations in other ships’ information and the need to collect all the other ships’ information to be systematically analyzed (Besikçi et al., 2016).

Among many practices available, the fuel-efficient operation is chosen as the focus of SEEMP development in this paper as it is easier to be adapted. Many factors can be considered for fuel-efficient operation such as improved voyage planning, which can be achieved using the help of different software tools, weather routing for specific routes and trade areas, just in time, or good communication with the port in order to give maximum notice of berth availability and facilitate the use of optimum speed to maximize efficiency and minimize delay, speed optimization while coordinating with the availability of loading or discharge berths, and optimized shaft power using the automated engine management system to control speed. There are other types of factor as well that could contribute to SEEMP and needs further research, such as optimized ship handling (trim, ballast, rudder, hull, and propulsion system), fuel type, and even the ship’s operational life service (MEPC, nd).

Machine learning is a method capable of analyzing the performance of existing systems, then determining the condition and improving the system performance to be more accurate. It could re-program computers when introduced to new information based on the initial learning strategies. The use of machine learning to develop green shipping industry has been done several times, including in determining the ideal ship fuel consumption for fuel efficiency and environmental preservation by conducting various training and tests on the ship’s machinery data to create an estimated model containing predictions that can be compared with actual data to produce the best model for ship fuel consumption optimization (Singh and Dhiman, 2021).

Previously, machine learning was used to predict the energy consumption needed by ships using gradient boosting regression (GBR), random forest (RF), BP network (BP), linear regression (LR), and k-nearest neighbor regression (KNN) processes, but the discussion of the strategy regarding ship energy consumption reduction is limited to the efficiency of port facilities and the arrival time of ships (Uyanik et al., 2020). Calculations on CO2 emissions and their distribution at ports are important to be investigated for SEEMP development (Peng et al., 2020). This research aims at helping to create decisions related to SEEMP development for ship owners and port authorities, which benefits green shipping activities, hence reducing CO2 emission and increasing shipping efficiency by looking at not only the total emissions but also the amount of CO2 emissions during the process of maneuvering, in the port basin, and berthing to make an analysis using machine learning models and a validation to replace the traditional estimation method, especially for container ports, which have not been widely researched.

Conclusion

    This paper uses the machine learning method to predict the variable importance of CO2 emission with RapidMiner Studio as the main tool to create a data model and analysis. A random forest model is created using five data characteristics as free variables. After running the model, variable importance is then selected and analyzed along with the effects of the selected variables on the ship’s CO2 emission. Finally, the SEEMP is developed based on the discussed result. Several options to minimize the CO2 emission during maneuvering, in the port basin, and berthing are discussed, using the results of the training model. While cold ironing has a huge CO2 emission reduction port, not all ports can provide it, so a combination between voyage optimization methods and trained operators is crucial for achieving the required EEDI reduction rates. Further research to minimize the CO2 emission by analyzing the port’s facilities, such as cold ironing feasibility and work efficiency improvement, need to be done, as cooperation with the port holds the crucial part during berthing processes, such as cold ironing availability and information exchange, to achieve just-in-time arrival. The method proposed by this paper can be used to develop fuel-efficient operations at ports for SEEMP development, especially container ports in developing countries such as Indonesia, where environmental health is not the top priority.

Acknowledgement

   This study is supported by funding for Basic Research from the Ministry of Research and Technology/National Research and Innovation Agency of the Republic of Indonesia for Fiscal Year 2021. Number: NKB-035/UN2.RST/HKP.05.00/2021.

References

Bergovist, R., Monios, J., 2018. Green Ports: Inland and Seaside Sustainable Transportation Strategies. Elsevier

Breiman, L., Friedman, J., Olsen, R., Stone, C., 1984. Classification and Regression Trees. Wadworth

Budiyanto, M.A., Shinoda, T., 2020. Energy Efficiency on the Reefer Container Storage Yard: An Analysis of Thermal Performance of Installation Roof Shade. Energy Reports, Volume 6(Supplement 2), pp. 686–692

Budiyanto, M.A., Huzaifi, M.H., Sirait, S.J., 2019. Estimating of CO2 Emission in a Cotainer Port based on Modality Movement in the Terminal Area. International Journal of Technology, Volume 10(8), pp. 1618–1625

Budiyanto, M.A., Huzaifi, M.H., Sirait, S.J., Prayoga, P.H.N., 2021. Evaluation of CO2 Emissions and Energy use with Different Container Terminal Layouts. Scientific Reports, Volume 11, pp. 1–14

Budiyanto, M.A., Pamitran, A.S., Wibowo, H.T., Murtado, F.N., 2020. Study on the Performance Analysis of Dual Fuel Engines on the Medium Speed Diesel Engine. Journal of Advanced Research in Fluid Mechanics and Thermal Sciences, Volume 68(1), pp. 163–174

Be?ikçi, E.B., Arslan, O., Turan, O., Ölçerc, A.L., 2016. An Artificial Neural Network Based Decision Support System for Energy Efficient Ship Operations. Computers & Operations Research, Volume 66, pp. 393­–401

Genuer, R., Poggi, J., Tuleau, C., 2008. Random Forests: Some Methodological Insights. Rapport de recherche, pp. 1­­–35

Huzaifi, M.H., Budiyanto, M.A., Sirait, S.J., 2020. Study on the Carbon Emission Evaluation in a Container Port based on Energy Consumption Data. Evergreen, Volume 7(1), pp. 97–103

Iris, C., Lam, J.S.L., 2019. A Review of Energy Efficiency in Ports: Operational Strategies, Technologies and Energy Management Systems. Renewable and Sustainable Energy Reviews, Volume 112, pp. 170–182

MARPOL, nd. Annex VI Chapter 4 Regulatio 21.6: Review of Phases and Reduction Rates

Mengxiao, L., Hu, S., Ge, Y., Heuvelink, G.B.M., Ren, Z., Huang, X., 2021. Using Multiple Linear Regression and Random Forests to Identify Spatial Poverty Determinants in Rural China. Spatial Statistics, Volume 42, https://doi.org/10.1016/j.spasta.2020.100461

MEPC, nd. 70/18/Add.1 Annex 10, 2016 Guidelines for the Development of a Ship Energy Efficiency Management Plan (SEEMP)

Pamitran, A.S., Budiyanto, M.A., Maynardi, R.D.Y., 2019. Analysis of ISO-tank Wall Physical Exergy Characteristic – Case Study of LNG Boil-Off Rate from Retrofitted Dual Fuel Engine Conversion. Evergreen, Volume 6(2), pp. 134–142

Peng, Y., Liu, H., Li, X., Huang, J., Wang, W., 2020. Machine Learning Method for Energy Consumption Prediction of Ships in Port Considering Green Ports. Journal of Cleaner Productions, Volume 264, https://doi.org/10.1016/j.jclepro.2020.121564

Mohana, R.M., Reddy, C.K.K., Anisha, P.R., Murthy, B.V.R., 2021. Random Forest Algorithms for the Classification of Tree-based Ensemble. Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2021.01.788

Sciberas, E.A., Zahawi, B., Atkinson, D.J., Juandó, A., Sarasquete, A., 2016. Cold Ironing and Onshore Generation for Airborne Emission Reductions in Ports. In: Proceedings of the Institution of Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment, pp. 67–82

Jensen, S., Lützen, M., Mikkelsen, L.L., Rasmussen, H.B., Pedersen, P.V., Schambya, P., 2018. Energy-Efficient Operational Training in a Ship Bridge Simulator. Journal of Cleaner Production, Volume 171, pp. 175–183

Singh, J., Dhiman, G., 2021. A Survey on Machine-Learning Approaches: Theory and Their Concepts. Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2021.05.335

Styhre, L., Winnes, H., Black, J., Lee, J., Le-Griffin, H., 2017. Greenhouse Gas Emissions from Ships in Ports – Case Studies in Four Continents. Transportation Research Part D: Transport and Environment, Volume 54, pp. 212–224

Tsagrisa, M., Pandis, N., 2021. Multicollinearity. American Journal of Orthodontics and Dentofacial Orthopedics, Volume 159(5), pp. 695–696

Uyan?k, T., Karatu?, Ç., Arslano?lu, Y., 2020. Machine Learning Approach to Ship Fuel Consumption: A Case of Container Vessel. Transportation Research Part D: Transport and Environment, Volume 84, https://doi.org/10.1016/j.trd.2020.102389

Weng, J., Shi, K., Gan, X., Li, G., Huang, X., 2020. Ship Emission Estimation with High Spatial-Temporal Resolution in the Yangtze River Estuary using AIS Data. Journal of Cleaner Production, Volume 248, https://doi.org/10.1016/j.jclepro.2019.119297

Witten, I.H., Frank, E., Hall, M.A., 2011. Data Mining: Practical Machine Learning Tools and Techniques. Elsevier Inc

Xing, H., Spence, S., Chen, H., 2020. A Comprehensive Review on Countermeasures for CO2 Rmissions from Ships. Renewable and Sustainable Energy Reviews, Volume 134, https://doi.org/10.1016/j.rser.2020.110222