Predicting Cargo Arrival Time Using Scala and Spark: Approaches and Achievements

Аннотация

The article examines methods to predict cargo PUBLISHED 12 March 2025 VOLUME Vol.07 Issue03 2025 arrival times through Apache Spark and Scala. The necessity for such methods arises due to external CITATION factors such as unpredictable road conditions, weather phenomena, and specific logistical operations. Danylo Liakhovetskyi. (2025). Predicting cargo arrival time using scala and spark: approaches and achievements. The American Journal of Information processing employs methods such as Engineering and Technology, 7(03),105–111. regression, decision trees, and neural networks, which https://doi.org/10.37547/tajet/Volume07Issue03-09 analyze data from sensors, GPS devices, and other COPYRIGHT sources to build forecasts that consider all factors © 2025 Original content from this work may be used under the terms directly or indirectly affecting calculation accuracy. of the creative commons attributes 4.0 License. The methodology is based on studying the functionality of the Apache Spark platform integrated with the Scala programming language, enabling the processing of large datasets with high operational speed and solution scalability. The use of Apache Spark combined with Scala accounts for streaming data, which improves prediction accuracy. This method optimizes logistics processes by reducing delays and allowing timely responses to changes in external conditions. The information presented in the article will be useful for data processing professionals, logisticians, and developers. Keywords: ETA, Scala, Apache Spark, logistics, machine learning, big data. Introduction: The need to predict the time of cargo delivery to its final destination is driven by the fact that arrival times directly impact the quality of transportation operations and the costs associated with these processes. Technological advancements have fundamentally transformed the process of estimating delivery times by The American Journal of Engineering and Technology 105 https://www.theamericanjournals.com/index.php/tajet The American Journal of Engineering and Technology leveraging machine learning technologies. These potential application of time series forecasting methods technologies are capable of processing large datasets using Apache Spark. The implementation of distributed directly sourced from various inputs, distinguishing systems is shown to allow for the efficient handling of them from traditional methods. The use of Apache large datasets, ultimately improving the quality of data Spark and Scala enables high-speed computations, analytics. which are essential for timely data processing and However, despite advancements in prediction accuracy, subsequent decision-making. unresolved challenges remain. Many studies focus The relevance of this topic is determined by the need exclusively on improving the precision of predictions, to process data obtained from various sources (such as without giving adequate attention to the usability of GPS or meteorological indicators). The implementation models or their adaptability to modern conditions. of Apache Spark or Scala allows for real-time responses Regarding the use of the Apache Spark platform and to ongoing changes and facilitates informed decision- Scala, researchers note that this combination allows for making. efficient data processing, ensuring the performance and scalability needed to address information analysis tasks. This study aims to analyze the opportunities that arise for companies through the application of Apache Spark The methodology of this study involves the use of the and Scala in logistics. Apache Spark platform with the Scala programming language, as this combination improves data processing speed and effectively addresses challenges related to