A new generation of smart stormwater systems promises to reduce the need for new construction by enhancing the performance of the existing infrastructure through real-time control. Smart stormwater systems dynamically adapt their response to individual storms by controlling distributed assets, such as valves, gates, and pumps. This paper introduces a real-time control approach based on Reinforcement Learning (RL), which has emerged as a state-of-the-art methodology for autonomous control in the artificial intelligence community. Using a Deep Neu- ral Network, a RL-based controller learns a control strategy by interacting with the system it controls - effectively trying various control strategies until converging on those that achieve a desired objective. This paper formulates and implements a RL algorithm for the real-time control of urban stormwater systems. This algorithm trains a RL agent to control valves in a distributed stormwater system across thousands of simulated storm scenarios, seeking to achieve water level and flow set-points in the system. The algorithm is first evaluated for the control of an individual stormwater basin, after which it is adapted to the control of multiple basins in a larger watershed (4 km 2 ). The results indicate that RL can very effectively control individual sites. Performance is highly sensitive to the reward formulation of the RL agent. Generally, more explicit guidance led to better control performance, and more rapid and stable convergence of the learning process. While the control of multiple distributed sites also shows promise in reducing flooding and peak flows, the complexity of controlling larger systems comes with a number of caveats. The RL controller’s performance is very sensitive to the formulation of the Deep Neural Network and requires a significant amount of computational resource to achieve a reasonable performance en- hancement. Overall, the controlled system significantly outperforms the uncontrolled system, especially across storms of high intensity and duration. A frank discussion is provided, which should allow the benefits and draw- backs of RL to be considered when implementing it for the real-time control of stormwater systems. An open source implementation of the full simulation environment and control algorithms is also provided.
Keywords: Real-time control | Reinforcement learning | Smart stormwater systems

MULTIDISCIPLINARY AND INTERDISCIPLINARY education can provide relevant insights into ubiquitous computing and other fields.1 In this article, we share our experience with multidisciplinary and interdisciplinary teaching in the twoyear Artificial Intelligence Research Master’s program at Utrecht University, the Netherlands. In particular, we zoom in on our motivation for, and experience with, revising courses in which nonengineering topics can be related to a more engineering inclined audience, and vice-versa.

The number of academic papers in the area of Artificial Intelligence (AI) and its applications across business and management domains has risen significantly in the last decade, and that rise has been followed by an increase in the number of systematic literature reviews. The aim of this study is to provide an overview of existing systematic reviews in this growing area of research and to synthesise their findings related to enablers, barriers and social implications of the AI adoption in business and management. The methodology used for this tertiary study is based on Kitchenham and Charter’s guidelines [14], resulting in a selection of 30 reviews published between 2005 and 2019 which are reporting results of 2,021 primary studies. These reviews cover the AI adoption across various business sectors (healthcare, information technology, energy, agriculture, apparel industry, engineering, smart cities, tourism and transport), management and business functions (HR, customer services, supply chain, health and safety, project management, decisionsupport, systems management and technology acceptance). While the drivers for the AI adoption in these areas are mainly economic, the barriers are related to the technical aspects (e.g. availability of data, reusability of models) as well as the social considerations such as, increased dependence on non-humans, job security, lack of knowledge, safety, trust and lack of multiple stakeholders’ perspectives. Very few reviews outside of the healthcare management domain consider human, organisational and wider societal factors and implications of the AI adoption. Most of the selected reviews are recommending an increased focus on social aspects of AI, in addition to more rigorous evaluation, use of hybrid approaches (AI and non-AI) and multidisciplinary approaches to AI design and evaluation. Furthermore, this study found that there is a lack of systematic reviews in some of the AI early adopter sectors such as financial industry and retail and that the existing systematic reviews are not focusing enough on human, organisational or societal implications of the AI adoption in their research objectives.
Keywords: artificial intelligence | business | machine learning | management | systematic literature review | tertiary study

Driven by the recent advances in industry 4.0 and industrial artificial intelligence, Automated Guided Vehicles (AGVs) has been widely used in flexible shop floor for material handling. However, great challenges aroused by the high dynamics, complexity, and uncertainty of the shop floor environment still exists on AGVs real-time scheduling. To address these challenges, an adaptive deep reinforcement learning (DRL) based AGVs real-time scheduling approach with mixed rule is proposed to the flexible shop floor to minimize the makespan and delay ratio. Firstly, the problem of AGVs real-time scheduling is formulated as a Markov Decision Process (MDP) in which state representation, action representation, reward function, and optimal mixed rule policy, are described in detail. Then a novel deep q-network (DQN) method is further developed to achieve the optimal mixed rule policy with which the suitable dispatching rules and AGVs can be selected to execute the scheduling towards various states. Finally, the case study based on a real-world flexible shop floor is illustrated and the results validate the feasibility and effectiveness of the proposed approach.
Keywords: Automated guided vehicles | Real-time scheduling | Deep reinforcement learning | Industry 4.0