Using a Big Data Analytics Approach to Unlock the Value of Refrigeration Case Parametric Data

Within the retail sector, refrigeration accounts from between 40% to 60% of a store’s total electrical energy budget where the most commonly used method in providing the necessary cooling energy to refrigeration cases is through multiplex direct expansion systems. Here all shop floor refrigeration cases use direct expansion air-refrigerant coils that are connected to banks of system compressors located in a remote machine room normally in the back or on the roof of the store, along with the supporting air-cooled condenser rooftop units. As a consequence of the highly regulated nature of the refrigeration process itself, driven by business legal requirements, and the need to maintain proper food quality levels, underlying fridge control parametric data is as a result, both very rich in content, and also readily accessible for compliance support reasons. And in data volume terms with individual cases being sampled typically every 5 minutes, this is estimated to generate upwards of 70-100 million discrete pieces of relevant control point data for a typical large size store, in a single year. Therefore this paper, through a series of examples from data taken for actual stores from a large retailer, explores the opportunities and value (economically and technically) of acquiring, harvesting, and applying big data aggregated statistical approaches to this large data set, to help the domain experts to deepen their knowledge of actual refrigeration case behaviour, and most importantly to gain greater understanding of the sources of variation in their underlying energy usage. It is shown that the follow-on energy savings from the knowledge gained from this data analytics approach can be significant, with one such project alone, relating to a defrost policy change, singularly capable of delivering over 2.5% saving in overall store energy usage. (Ref 1 : ComputerScope Article 2013 ). It is further presented, that through a set of follow-on developed engineering driven KPI’s taken from this readily available parametric data set, that this new insight will not only allow for additional energy saving through real-time case anomaly detection, but also has the potential to positively impact on the direction of future maintenance support models, supplementing traditional preventative/reactive methods with cost effective data analytics driven predictive maintenance approaches.

By: Niall Brady, Paulito Palmes, John Walsh

Published in: RC25443 in 2014


This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.


Questions about this service can be mailed to .