Learning and management skills: the internationalization of Grupo Arcor. Analysis from a process-based approach

This paper analyzes the path of the international expansion of Grupo Arcor, an Argentine multinational company specializing in confectionery. The objective is to entify corporate strategies and business learning that led this Latin American firm to establish itself as one of the leading manufacturers in confectionery industry ,particularly in the 21st Century. The analysis is primarily qualitative in order to identify the economic dimension as a determinant in the internationalization process; a processbased approach from the Uppsala Model is used for this. However, the study is also complemented with a regression analysis to test if the firm was driven to expand internationally by the expectations on the degree of globalization of the industry and the accumulation of experience in foreign markets, and if the company was influenced by psychic distance in choosing the location of its investment; given the influence of these variables in Grupo Arcor business strategies. Our findings suggest that Grupo Arcor, was able to become global due to strategies such as vertical integration, diversification of products and geographical markets (based on psychic distance) and indeed some strategies were consequence of the globalization of the sector and the accumulation of experience in foreign markets.

Learning Multirobot Hose Transportation and Deployment by Distributed Round-Robin Q-Learning

Multi-Agent Reinforcement Learning (MARL) algorithms face two main difficulties: the curse of dimensionality, and environment non-stationarity due to the independent learning processes carried out by the agents concurrently. In this paper we formalize and prove the convergence of a Distributed Round Robin Q-learning (D-RR-QL) algorithm for cooperative systems. The computational complexity of this algorithm increases linearly with the number of agents. Moreover, it eliminates environment non sta tionarity by carrying a round-robin scheduling of the action selection and execution. That this learning scheme allows the implementation of Modular State-Action Vetoes (MSAV) in cooperative multi-agent systems, which speeds up learning convergence in over-constrained systems by vetoing state-action pairs which lead to undesired termination states (UTS) in the relevant state-action subspace. Each agent's local state-action value function learning is an independent process, including the MSAV policies. Coordination of locally optimal policies to obtain the global optimal joint policy is achieved by a greedy selection procedure using message passing. We show that D-RR-QL improves over state-of-the-art approaches, such as Distributed Q-Learning, Team Q-Learning and Coordinated Reinforcement Learning in a paradigmatic Linked Multi-Component Robotic System (L-MCRS) control problem: the hose transportation task. L-MCRS are over-constrained systems with many UTS induced by the interaction of the passive linking element and the active mobile robots.