Improvement of logistics automation: a focus on unconventional solutions

Abstract

The benefits coming from logistics and warehousing are unanimously considered of paramount importance in industrial environments. A well-designed logistics allows important advantages, such as a reduced risk of stock-out, a mitigated bullwhip effect, and a reduced lead time. Moreover, the spread of automation which has taken hold in the last years can lead to further advantages (i.e. labour saving, greater specialisation of the employees, increased storage capacity, reduced throughput time, reduced recurrence of errors and damages, etc.). For this reason, during the last years, the scientific community focused on the implementation of automated solutions in the logistics field, addressing several design, management and control issues. Many automated solutions have been studied, although it seems that some industrial fields have been neglected. One of them is the steel industry, which is characterised by a high-level automation in production and manufacturing tasks, boasts over USD 500 billion value-added per year, and employs more than 6 million people. Nevertheless, it is poorly studied from the logistic point of view, and the automated storage and retrieval solutions are few and little studied in relation to its importance in the global economy. The objective of this work is to fill this gap by proposing several algorithms, methodologies, and operative policies focused on this neglected sector to improve different aspects of logistics. More in detail, three different crucial aspects such as (i) logistics of small parts, (ii) logistics of bulky parts, and (iii) managerial issues are considered, and, for each of them, the most widespread problems are addressed. In doing this, not only control policies and operational aspects are considered, conversely, the implementation of unconventional automated storage and retrieval solutions are analysed and their functioning is improved by proposing new algorithms, design decisions and control policies. Each time a new algorithm, operating policy, or methodology is proposed, a case study is carried out to validate it in a real industrial case, or, alternatively, comparing it to the solutions already proposed in literature.