A data-driven customer complaint management model for residential building companies

ABSTRACT

Many companies in charge of the development and construction of residential building projects do not appropriately manage customer complaint records, especially regarding providing feedback about quality. Data collection and procedures for data analysis are often ineffective, which limits the generation of knowledge. Previous studies on customer complaints have been mostly focused on the analysis of large databases but do not propose overall solutions for managing this type of information, including data collection and analysis. This investigation aims to devise a data-driven customer complaint management model for providing feedback to the design and production phases of residential building projects. This model has a set of protocols for: (i) data collection that can be used for developing digital applications; and (ii) analyzing defect cause–effect relationships, considering the experts’ perspective, by using Directed Acyclic Graph and Bayesian Network. This investigation was conducted in collaboration with a Brazilian residential building company, using Design Science Research as methodological approach. The main outcome of this investigation is a set of data collection and analysis protocols that support the migration from the traditional approach of simply providing repair services during the defect liability period to a data-driven customer complaint management approach that provides feedback to quality management systems.

HIGHLIGHTS

• A customer complaint management model must consider several perspectives;

• Digital applications must be used to support the implementation of data collection on defects;

• A complaint management model must deal with the complexity of defect formation.

Abbreviations: BN: Bayesian Network; CRM: Customer Relationship Management; DAG: Direct Acyclic Graph; DLP: Defect Liability Period; DSR: Design Science Research; MEP: Mechanical, Electrical, and Plumbing Systems

KEYWORDS:

• Complaints
• defects
• causes
• residential building
• Bayesian Network

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Council for Scientific and Technological Development through the Academic Doctorate for Innovation Program [grant number 142267/2019-8].