LINE-TRACK: technology to improve overall yield during the manufacturing process

Lead Research Organisation: Cranfield University
Department Name: School of Water, Energy and Environment

Abstract

The project aims to devise innovative technologies and engineering solutions in food & drink manufacturing. Process yield
losses are currently identified by the mass balance method; at the end of the production run the output is compared to the
input of the various process ingredients and materials. This method only identifies the losses after the event and does not
determine where the losses have occurred and so required improvements are difficult to identify accurately. The next
product run materials may be different and so the identified improvements may not be fully applicable. The proposed
tracking technology called LINE-TRACK adopts the principles of prognostics to measure the yield losses in real time. LINETRACK
can be used to 1) identify the points where losses occur; 2) help identify root causes of the losses; 3) alert the
operators as soon as the losses are above specified targets & 4) directly intervene in the system with the ultimate goal to
avoid losses occurring.

Planned Impact

ECONOMIC. CCE will benefit from cost avoidance in increased yield with decreased effluent and energy. LVS improves
sustainable growth with a new product which can be added to existing products, and by selling LINE-TRACK to overseas
markets, it will give positive impact on UK GDP and trade deficit. LINE-TRACK will open up business opportunities beyond
food and drink sector, combining the strength of LVS and Cranfield.
SOCIAL. For 51 years CCE Sidcup has been a significant manufacturing facility for the local and national manufacturing
community with a current workforce of 333 employees. Being competitive whilst ensuring a sustainable business for the
next 50 years will allow CCE to support the social infrastructure of the Sidcup local community, job security, better work
environment and a strong contributor to UK manufacturing output.
ENVIRONMENTAL. The project gives benefits in terms of reduction of effluent, energy and packing losses and ultimately
the carbon footprint. We envisage that the benefits can be achieved within 1 year of the project execution.
 
Description The project aims to devise innovative technologies and engineering solutions in food and drink manufacturing to reduce losses. Losses in manufacturing typically occur due to failures in the processes and the operators react to those failures adopting a fail and fix approach instead of prevention. The project will challenge existing practices applying a proactive methodology by adopting the principles of prognostics to predict failures in the targeted areas in real time. LINE-TRACK will be used to 1) identify the points where failures occur; 2) help identify root cause(s) and symptoms of the failure; 3) alert the operators that a failure is likely to occur & 4) directly intervene in the system with the ultimate goal to avoid failure(s) occurring or eliminate unplanned stoppages.
Exploitation Route The direct adopters of this project are Accolade Wines (AW) and LineViewSolutions (LVS). AW will use LINE-TRACK to improve their sustainability and competitiveness through increased efficiency of their manufacturing facilities. LVS improves sustainable growth with a new product which can be added to existing products, and by selling LINE-TRACK to overseas markets, it will give positive impact on UK GDP and trade deficit. LINE-TRACK can potentially be adopted by all manufacturers who use high speed lines to fill containers.
Sectors Agriculture, Food and Drink,Manufacturing, including Industrial Biotechology
 
Description The LINE-TRACK project has enabled the food and drinks manufacturing industry to adopt the principles of prognostics and condition monitoring in order to pinpoint the source of losses, so as to avoid production/yield losses. The predictive algorithms developed in this project predict failures of equipment in real-time and determine when the failure will likely occur, enabling the manufacturing system to stop production and conduct the maintenance regime with minimal disruption to the entire system. The principle of prognostics and condition monitoring can also be extended and applied to other sectors with similar characteristics and manufacturing control systems. The algorithms have been handed over to the industrial partner for further exploitation and commercialisation.
Sector Agriculture, Food and Drink,Manufacturing, including Industrial Biotechology
Impact Types Societal,Economic
 
Description Improving efficiency of food manufacturing sector
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Influenced training of practitioners or researchers
Impact This was done through a research seminar at Cranfield and other universities. The impact may not be necessarily be in terms of change of policy, but it is hoped that the it will reach the practice. The research seminar was aimed at food and drink practitioners, equipping them with awareness of emerging technology that might be relevant to their sector.
 
Description Assessment of the impacts of changes in packaging to upstream production in beverage industries
Amount £10,000 (GBP)
Organisation The Coca-Cola Company 
Sector Private
Country United States
Start 05/2013 
End 12/2013
 
Description Yield improvement at a bottling plant
Amount £10,000 (GBP)
Organisation The Coca-Cola Company 
Sector Private
Country United States
Start 04/2014 
End 12/2014
 
Title Skid filter algorithm 
Description The model intends to predict the remaining useful life of the filters based on the two major failures which were identified during the research. It includes some elements of machine learning to improve future predictions by using a similarity-based interpolation approach. The work is expected to stimulate future development of prognostics applications in multiple processes in the food and drink sector and to be beneficial to other companies with similar processes. 
Type Of Material Data analysis technique 
Year Produced 2015 
Provided To Others? Yes  
Impact This is the first attempt in the development of a predictive model within the Linetrack project and this has formed a basis for the development of the next model which will be looking at the parameters of the bag making machine. The underlying principle remains the same, i.e. exploring the application of prognostics techniques to reduce the faulty bags. 
 
Description AW and LVS 
Organisation Accolade Wines
Country Australia 
Sector Private 
PI Contribution This is on top of the contributions stated in the research project. The expertise of prognostics and condition-based monitoring in process industry has been transferred to staff of AW and LVS.The contributions are manifested through an MSc project where the PI acts as the academic supervisor.
Collaborator Contribution AW and LVS provided avenue for industrial research and exposure to real life projects.
Impact A paper is currently being written for a possible publication in a peer-review journal.
Start Year 2014
 
Description AW and LVS 
Organisation LineView Solutions
Country United Kingdom 
Sector Private 
PI Contribution This is on top of the contributions stated in the research project. The expertise of prognostics and condition-based monitoring in process industry has been transferred to staff of AW and LVS.The contributions are manifested through an MSc project where the PI acts as the academic supervisor.
Collaborator Contribution AW and LVS provided avenue for industrial research and exposure to real life projects.
Impact A paper is currently being written for a possible publication in a peer-review journal.
Start Year 2014
 
Title Skid filter 
Description Our role in this project is to develop the algorithm and prognostics models. These models have been handed over to the exploitation partners, Lineview Solutions. They will turn the models into software tools. 
Type Of Technology Software 
Year Produced 2015 
Impact Potentially, the software can be embedded into Lineview's product portfolio and be customised for other products. At this stage, this can only be corroborated by them.