Design and modeling of a waiter robot

The science of robotics is considered one of the most practical sciences in all fields. The application of this science is visible in all kinds of work fields and related fields, from construction activities to activities in the fields of medicine or even social services. One of the social services that are very widely used, is delivering items and orders to customers. This work is the duty of people who are called waiters. This job has very few benefits for people working in this field. Also, things like illness can cause some delay in the employer's work or not complete his work in some cases, also in situations such as when contagious diseases have spread, the direct communication of people within a short distance can cause more spread of the disease. The devices ordered by the customers could increase the speed of work and have a low-risk connection, the costs of the employer could be reduced, perfect service could be given to the customers, and the workforce could be employed for more useful work. This robot is specifically designed to use for reception in the conference hall of the growth center of Kharazmi University to receive the people present in this conference hall, but as mentioned above, these robots can be used in other places such as hospitals for delivering medicines to patients, also can be used in restaurants to deliver customer's orders to them. With this replacement, the speed of catering increases, at the same time, there is no lack of accuracy, and the issue that becomes more important with the spread of the contagious disease Covid-19 is hygiene, which can achieve several important goals in this field with this replacement. Specifically, during the reception, the distance between the host and the guest is less than one meter and is unsafe. Also, there is a possibility that each of the parties is a carrier of contagious diseases, and these problems are solved by this replacement. © 2022 IEEE.

Dynamic Inventory Replenishment with Reinforcement Learning in Managing E-Fulfilment Centres

The demand of retail e-commerce has been rapidly growing due to the digitalization and the COVID-19 pandemic, and thus, the stress on e-fulfilment services continues to increase nowadays. To fulfil daily customers’ orders, effective inventory replenishment is of the essence in order to strike a balance between inventory management costs and service level. This paper describes an enhanced inventory replenishment approach by using reinforcement learning to deal with non-stationary and uncertain demand from customers. The proposed approach relaxes the assumption of stationary demand distribution considered in typical inventory models. Conventional policies derived from such models cannot guarantee optimal re-order quantities, when demand distribution is non-stationary over time. Consequently, reinforcement learning is adopted in the proposed approach to improve feasible solutions continuously in a dynamic business environment. In comparison to the conventional base stock policy, our proposed approach provides cost saving opportunities ranging from 28.5 to 41.3% in a simulated environment. It is found that the value of using data-driven solution approaches to deal with the practical inventory management problem is effective. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Order Assignment and Scheduling for Personal Protective Equipment Production During the Outbreak of Epidemics

This paper investigates a new multi-objective order assignment and scheduling problem for personal protective equipment (PPE) production and distribution during the outbreak of epidemics like COVID-19. The objective is to simultaneously minimize the total cost and maximize the PPE supply timeliness. For the problem, we first develop a bi-objective mixed-integer linear program (MILP). Then an epsilon-constraint combined with logic-based Benders decomposition method is proposed based on some explored properties. We then extend the proposed model to handle dynamics and randomness. In particular, we design a predictive reactive rescheduling approach to address random order arrivals and manufacturer disruptions. Computational experiments on a real case from China and 100 randomly generated instances are conducted. Results show that the proposed algorithm significantly outperforms an adapted epsilon-constraint method combined with the proposed MILP and the widely used non-dominated sorting genetic algorithm II (NSGA-II) in obtaining high-quality Pareto solutions. Note to Practitioners-The unprecedented outbreak of COVID-19 and its rapid spread caught numerous national and local governments unprepared. Healthcare systems faced a vital scarcity of PPEs. The urgency of producing and delivering PPEs increases as the number of infected cases rapidly increases. A key challenge in response to the epidemic is effectively and efficiently matching the demands and needs. Performing practical and efficient order assignment and scheduling for PPE production during the COVID-19 outbreak is critical to curbing the COVID-19 pandemic. This work first proposes a bi-objective mixed-integer linear program for optimal order assignment and scheduling for PPE production. The aim is to achieve an economical and timely PPE production and supply. A novel method that combines the epsilon-constraint framework and the logic-based Benders decomposition is proposed to yield high-quality Pareto solutions for practical-sized problems. Computational results indicate that the proposed approaches are practical and feasible, which can help decision-makers to perform acceptable order assignment and scheduling decisions.