This paper presents a real application of a lean–green improvement initiative conducted at a large Portuguese hypermarket store. It explores how lean tools and techniques may be used to not only improve the operational performance, but also sustainability. A case study was carried out in one store of a multinational retail enterprise, with the aim of enhancing both the operational and sustainability performance in the cold meat section, one of the most relevant areas of the fresh food markets. The Gemba Kaizen event approach, which comprises three main stages, was adopted. During the workshop stage, the structured problem-solving methodology was followed, and was recorded in an A3 format. As a consequence of this project, food waste in the cold meat market was reduced by half, whereas the out-of-stock index decreased by a third. In addition, the pilot store hit top performance within all stores of the company in Portugal, ranking first in all key indicators for the cold meat market. The lean–green scope and performance improvement procedures developed and implemented in the pilot store were later deployed to other stores of the company. This is one of the first publications regarding the application of lean management in the food retail sector for improving both the operational and sustainability performance.

This paper describes how we tackled the development of Amaia, a conversational agent for Portuguese entrepreneurs. After introducing the domain corpus used as Amaia’s Knowledge Base (KB), we make an extensive comparison of approaches for automatically matching user requests with Frequently Asked Questions (FAQs) in the KB, covering Information Retrieval (IR), approaches based on static and contextual word embeddings, and a model of Semantic Textual Similarity (STS) trained for Portuguese, which achieved the best performance. We further describe how we decreased the model’s complexity and improved scalability, with minimal impact on performance. In the end, Amaia combines an IR library and an STS model with reduced features. Towards a more human-like behavior, Amaia can also answer out-of-domain questions, based on a second corpus integrated in the KB. Such interactions are identified with a text classifier, also described in the paper.

No abstract

No abstract

No abstract