Page 360 - ISC PROCEEDINGS 21.4

In logistics and supply chain management, computer-based analytical systems have
long been widely applied. Supply chain planners often rely on software tools to analyze
historical data for demand forecasting, while enterprise resource planning (ERP) systems
automate decision-making processes and warehouse management systems (WMS)
optimize storage and distribution operations. These digital tools can operate
independently or integrate with other business functions such as finance and supplier
management, enabling information sharing across a unified platform and improving
coordination efficiency. When connected to the internet, web-based systems allow
remote access and integration with third-party applications through APIs. Additionally,
cloud computing provides scalable IT infrastructure to support both short-term and long-
term computational needs, becoming a key component of modern supply chain systems
[3], [4].
Recent advances in logistics digitalization are driven by the integration of physical
assets with real-time data streams. Equipment and vehicles are monitored through
sensors that continuously collect operational data, while mobile devices can be used to
update information when necessary. This interconnected environment is a defining
feature of the industry 4.0 revolution. As a result, logistics operators gain near real-time
visibility across supply chain activities. Digital control towers, similar to air traffic control
systems, can detect potential disruptions such as inventory shortages before they occur.
With the support of algorithms and historical data, predictive analytics can identify
patterns and relationships that may not be easily recognized by human analysts. Real-
time analytics and optimization tools further support decision-making by generating
actionable recommendations, enabling managers to respond more effectively to
operational challenges [13], [14].
Despite these advancements, AI does not replace human decision-making. Instead,
it enhances managerial capabilities by providing predictive insights. In most current
logistics applications, human managers still retain the final authority, ensuring that
decisions align with strategic and contextual considerations.
2.2. Smart logistics
The application of digital technologies, combined with sensor-based asset
connectivity and digital control towers, generates an enormous volume of real-time data
within logistics operations. The key challenge is how to effectively utilize this data to
enhance intelligence in logistics operations and supply chain decision-making.
The use of data in logistics is not a new concept. For decades, global trade has relied
on forecasts and historical information to support the movement of goods across
international markets. However, the distinguishing feature of the current digital era lies in
the scale and speed at which data are generated, stored, and exchanged, enabling
logistics and supply chain management systems to become significantly more flexible and
intelligent.
In traditional data-driven logistics systems, decision-making typically relied on a
limited number of data sources, such as historical demand patterns or current inventory
levels. Based on quantitative inputs, rule-based systems could be programmed to support
or automate operational decisions. In contrast, modern digital logistics environments
involve massive datasets generated from numerous sources.
These datasets are collected automatically through Internet of Things (IoT) sensors
and manually through mobile and handheld devices, often referred to as the Internet of
People. As the number of data sources increases exponentially, the complexity of data

359