Bayesian networks as a guide to value stream mapping for lean office implementation: a proposed framework

Bayesian networks (BNs) are recognized worldwide for their ability to work with reasoning involving uncertainty. BNs are formed by a directed acyclic graph whose nodes are random variables with different states with associated conditional probabilities. BN allows identifying and prioritizing where to act first and simulating its results to avoid costs with changes without significant results. On the other hand, the lean office (LO) philosophy is recognized for focusing on reducing waste (which does not add value to the product or service) applied to the office environment. One of its main tools is the value stream mapping (VSM), which assists in the lean transformation, aiming at the visualization and elimination of wastes. In this context, this paper aims to propose a framework for using BNs as a guide in the VSM elaboration to implement the LO, with prioritization of lead time (LD) reduction more reliably. To evaluate the validity of the proposed framework, a case study was conducted in a product development department of an electronics industry. The results demonstrated the feasibility and effectiveness of the proposed framework. Accordingly, the contributions of this paper are twofold. In theoretical terms, it promotes increased knowledge by exploring the combination of BNs with LO. In practical terms, the proposed framework is easy to apply and understand, allowing managers and professionals to implement it to reduce lead time in other types of processes in different industries. Thus, it can support decision-makers in eliminating waste in their processes.

Automated summary

This paper proposes a framework that combines Bayesian networks with lean office to guide the implementation of lean office through value stream mapping, with a more reliable prioritization of lead time reduction. A case study conducted in a product development department validates the feasibility and effectiveness of the proposed framework. The paper contributes to increased knowledge by exploring the combination of Bayesian networks and lean office in theory, and provides an easy-to-use and understandable tool for managers and professionals in different industries to reduce lead time in practice.