prof.dr. M.E. Iacob (Maria)

Big Data for Resilient Logistics

DataRel

The modern supply chain continues to seek more cost savings and greater transparency and efficiency in all processes across the whole chain. The Internet of Things (IoT) is fundamentally transforming the transport industry and logistics. Next- generation smart logistics will optimize the movement of people and goods locally and globally, improving economics, public safety, and sustainability. This requires a multi-tiered intelligent system architecture providing high degree of modularity and autonomy, satisfying demanding and sometimes conflicting requirements of smart logistics and transport in terms of scalability, availability, and security. The wide availability of sensors (e.g., smart packaging, sensing technology, temperature control, board computers, GPS trackers, etc.) in logistics supply chains and the wide-band communication of Internet of Things (IoT) allow the collection of Big Data from sensors placed at key points in the logistic supply chain. This project is a collaboration between several companies and universities, and partially funded by NWO. Within this project we are focusing on the following areas: - Logistics Internet of Things (Pervasive Systems Group) This PhD research will focus on a new approach to logistic processes involving cyber-physical networks of logistic entities (e.g., goods, vehicles, infrastructure) that embed smart sensors and business logic. These entities, called Smart Returnable Transport Items (SRTIs), are capable of collecting data from extant IoT devices and social networks. The main goal is to detect unexpected behavior and do adaptive sampling of data generated by collaborating SRTIs in the region of interest, in order to solve problems such as, lost and damaged perishable goods during transportation, waste reduction, handling, and storage throughout the lifecycle of products based on sensing and location capabilities, etc. - IoT-driven resilient multimodal planning in smart logistics (Industrial Engineering and Business Information Systems Group) This PhD research will focus on a new approach to use this data input provided by cyber-physical networks of logistic entities (e.g., goods, vehicles, infrastructure) that embed IoT devices in the real-time multi-modal planning of orders. More concretely, the main goal is to improve the resilience and performance of planning decisions in terms of sustainability, efficiency, queue delay, costs and quality, by developing Artificial Logistics Intelligence (ALI) capable of detecting and dealing with emergent phenomena (given large volumes of heterogeneous data) and capable of learning from the experience of human planners using machine learning.

Industry 4.0 driven Supply Chain Coordination for Small- & Medium-sized Enterprises (ICCOS)

ICCOS aims at improving the competitiveness of the Dutch logistics sector by increasing the adoption and usage of industry 4.0 related technologies in combination with advanced real-time data analytics. ICCOS proposes a Logistic Industrial Data Space (LIDS) architecture with applications in logistics decision making and planning.

CLICKS IDS CONNECTOR STORE AND INTEROPERABILITY SIMULATOR FOR SMEs

This project develops the Logistic Data Space (LDS), which allows logistics companies to participate in digital ecosystems for efficient collaboration and new services. LDS consists of a Connector Store, offering connectors to realize data exchange between heterogeneous ICT environments, and an Interoperability Simulator, to explore collaboration opportunities prior to implementation.

Defining, Designing, and Implementing Rural Smartness. The case of West Java

This project has three main research goals: RG1: To understand the causal mechanisms underlying the realisation of rural smartness and subsequently leading to economic welfare improvement for rural citizens. For this purpose, we develop a theoretical model that depicts the interplay between the determinants of rural smartness and its implications on the economic welfare of rural citizens. Knowledge on the social artefact of rural smartness explained by this model will provide practical guidance for the implementation of initiatives toward rural smartness and allow us to identify the more accurate requirements of the IT artefacts supporting rural smartness. RG2: To design a reference architecture that could guide the provision of rural smartness platform: a digital service platform that facilitates the establishment of a Digital Business Ecosystem (DBE) tailored to improve the rural economic climate. We provide the reference architecture by means of an enterprise architecture design using the ArchiMate modelling language. The reason for using this approach is that enterprise architecture allows for a multi-layered approach to modelling, and ArchiMate is expressive enough to allow the explicit modelling of services and service exchanges. Furthermore, the design of the reference architecture is grounded in the theoretical model mentioned in RG1. RG3: To achieve the intended societal impact, i.e., the improvement of rural citizens’ economic welfare, the reference architecture needs to be implemented in a real setting. To provide guidance on how the reference architecture could be translated into the actual IT artefact, i.e., the rural smartness platform, we follow the technical action research (Wieringa & Morali, 2012) methodology, by working closely with the ICT agency of the provincial government of West Java, Indonesia. Furthermore, to understand how the rural communities (as the intended target group) are willing to use the platform as part of their business activities, we conducted an empirical study by means of a survey with respondents from rural areas of West Java.

Reinforcement LeArning Platform for SMEs in Logistics (ReAL)

While real time data and sophisticated (deep) RL approaches are emerging, logistic organizations (in most cases SMEs) lack the tools and expertise to effectively identify whether (parts) of their business processes are suitable for reinforcement learning and adopt these state of the art research in their daily practice. Taking as starting point the results we booked in this area in our previous projects (ICCOS, and Autonomous Logistic Miners), the ReAL project combines the ideas behind the hybrid intelligence paradigm and combines RL with a common data model and industry standard (such as the Open Trip Model) in order to achieve data and process interoperability and support a broader range of logistics decision making processes. ReAL aims to (i) Design, develop, test and implement a RL-based decision support platform based on the Open Trip Model (OTM) and accessible to logistic SMEs. (ii) Complement the ReAL platform with teaching material and training to stimulate its dissemination of throughout the logistic SME community. ReAL will cover the following main functional components: a main common OTM-based data model, a tool for testing multiple RL algorithms and perform (hyper)parameter tuning, and an infrastructure for the management, execution and monitoring of RL agents. ). The platform will be complemented by training materials to transfer the platform, knowledge, and practices to the SMEs via Logistics Associations. ReAL contributes to the data-driven logistics theme by combining advanced data analytics, reinforcement learning and OTM as part of the Federated Data Sharing Infrastructure for the Dutch Logistics sector.

Smart Circular Construction Ecosystems

The research objective of this project is twofold: 1. To propose a Circularity Information Platform (CIP) reference-architecture encompassing functionality, such as CDW-information-sharing, collaborative negotiation and decision-making support in the form of agent-based simulation framework, business analytics, circular supply chain design and coordination; 2. To develop and validate a bi-directional regulatory framework that realizes both the translation of high-level policies into actionable business rules (encapsulated in the CIP’s coordination tooling), and the elicitation (using CIP’s analytical tooling) of specific emergent systematic behaviour and industry requirements concerning a tailored Circular Economy policy-making process.