A Project to Develop and Introduce Automated, Digital Imaging Technology to the Grading Process in Canadian Mussel Plants

Final Report

Confederation Cove Mussels Ltd.

AIMAP 2011-G02

Executive Summary

Confederation Cove Mussels Co Ltd (Conference Cove) is a recognized leader in the mussel processing industry in Prince Edward Island. This innovative company recognized the stagnant rate at which new technology has been introduced to the mussel processing industry. To remain competitive in the global marketplace, Confederation Cove has keenly promoted the introduction of new, automated techniques to reduce the heavy dependence on labor and to improve the product quality levels that can be achieved. The challenge for processing plants has been the inability to find research and development funding models that would be attractive enough for companies to pursue the expensive process. The Sustainable Production Component of the Aquaculture Innovation and Market Access Program (AIMAP) of the Department of Fisheries and Oceans (DFO) provided that vehicle.

The intent of this project was to develop an innovative grading system that would separate mussels for Size, Shape and Defects. The equipment would use the latest technology in image acquisition and analysis using specially designed software algorithms to grade mussels at a high speed. The original design focused on mussels being fed over four grading belts with their images acquired in real time to allow grading at a rate of eight mussels per second per lane or 25-30 mussels per second per unit. This would allow commercial production rates that make this automated process of mussel grading commercially attractive. The application of this technology to other complex operations would be tested. This would include mussel separation at high speed for meat content while also identifying and removing any defective units.

The project has been completed and the key deliverables have been achieved. The equipment has been fabricated to a food grade standard and it effectively grades mussels for size and identifies and removes defective units. It operates at a speed and performance rating that make it an attractive option for any mussel processor.

However, like any research and development project of this type, there were significant challenges to be overcome along the way. The digital imaging equipment developed by Lizotte Consultants required a significant amount of development work to allow it to grade the mussels effectively. The greatest challenge to be overcome was the need to custom design the mussel handling equipment required to feed the digital imaging equipment and the post-imaging equipment used to remove and handle the graded mussels. Without the support of AIMAP, it would have been very difficult to achieve this level of success with this project. 

1.0 Project Introduction

This report is meant to summarize the work and outcomes associated with the “Project to Develop and Introduce Automated, Digital Imaging Technology to the Grading Process in Canadian Mussel Plants”. The project was funded by the Sustainable Production Component of AIMAP of the Department of Fisheries and Oceans.

This project was designed to address the lack of technology currently employed in the grading process in the mussel processing industry. Most success stories in innovation have continuously demonstrated that the most important managerial strategy for maintaining a competitive position in the global economy is the introduction and maintenance of new technology. Most productivity improvements in manufacturing are associated with the introduction of innovative equipment that lowers labor costs or increases outputs.

The new developments in imaging technology using customized software algorithms provided this project with a tremendous opportunity to not only automate the mussel grading process but to deliver new levels of applications that would be unimaginable in years past. The new systems are based on digitizing the images to be managed and then analyzing them on the parameters of concern such as size, shape, colour and defects.

The outcome of this analysis are manifested in process control signals that actuate corresponding processes, including rejection, selection, sorting and grading. Advantages of these systems are manifold, including consistency and efficiency. This report will deal with the scope of that project.

The key objectives of this report are to:

• Provide insight into the challenges that were addressed through the research.
• Indicate the methodology that was employed to address the issue.
• Identify the deliverables that were achieved.
• Identify the technical challenges that had to be overcome.
• Provide insight into the timelines and milestones involved.
• Recommend initiatives that could be explored to improve future research.
• Support the Aquaculture Innovation and Market Access Program and its role.

The report will provide an evaluation of the success of the project relative to its proposed objectives. It will also provide some insight into lessons that might be applied to other projects such as this in the future.

2.0 Project Methodology

Project Challenge

The challenge and the opportunity provided involved in this project related to the mussel grading process and its labor intensive nature. The mussel processing industry is a relatively new business and, up until this project, the grading activity has remained a completely manual process. Workers would stand along a conveyor and manually assess and visually grade the product to ensure that defective mussels do not make it to the packaging process. It is an expensive and inconsistent process.

This grading technique presents several challenges that limit the industry's ability to continuously improve its process and respond to ever increasing marketplace challenges. These challenges include added production costs, the difficulty in attracting dependable labor and the natural inconsistency in the mussel grading performance.

There were attempts at introducing grading equipment in the past but there was no measure of commercial success in solving the grading challenge. The intent of this project was to design and develop an automated grading machine that would consistently:

• Identify, select and remove cracked, broken, empty, or full of full of mud mussels
  
• Identify, select and remove common foreign material found with mussels
  
• Grade mussels for their size
  
• Grade mussels for their general meat content
  
• Perform the above four activities at least to a 95 % confidence level
  
• Perform the above activities at a speed of 5,000 to 6,000 lbs of mussels per hour

While there has been at least one attempt in the past to provide this type of technology, a commercial mussel grader that is anywhere close to the above performance specifications does not exist.

Project Design

Confederation Cove identified and recruited Lizotte Consultants, a New Brunswick company, to help deliver this project. Lizotte Consultants have experienced significant success in introducing high speed, very reliable on-line inspection equipment. Their applications range from 100 % on-line inspection of cans in seafood to 100 % on-line inspection of board lumber. All participants were aware that an introduction of this type of technology to the mussel processing industry would create a paradigm shift in terms of labor costs and quality performance but would also present difficult technical challenges.

The challenge presented to Lizotte Consultantswas to custom designa system that would grade mussels for Size, Shape, Defects and Meat Content.This would be a new manufacturing system that would use image acquisition and analysis using a specially designed software algorithm to grade mussels at a high speed. The original concept focused on mussels being fed over four grading belts with their images acquired in real time. The goal was to allow grading at a rate of 8 mussels per second per lane or up to 32 mussels per second per unit. This would allow commercial production rates that make this automated process of mussel grading very attractive. However, this new technology would be developed to allow more complex operations to be conducted. The mussels would be separated at high speed for such elements as size, shape and meat content while also identifying and removing any defective units. 

Confederation Cove was also acutely aware that the development of the Auxiliary Handing Equipment to interface the automated grader to the existing line would be a challenge. To ensure success Atlantic Systems Manufacturing (ASM) of Charlottetown were recruited to the research and development team.

Project Methodology

The principals at Confederation Cove were well aware that a research and development project of this nature would be full of pitfalls and that the road map would have many unexpected turns. That is exactly what occurred and is still occurring with this project. The Project Management Plan involved the interfacing of several key parties. This was important to ensure effective project communication. The key positions were:

1. The lead role was provided by the Project Manager who coordinated all activities and specifically the delivery of the Work Plan.
   
2. The Project Manager worked very closely with the Project Engineering Consultant which was filled by Atlantic Systems Manufacturing.
   
3. The critical interface with Confederation Cove was provided by the General Manager and his Project Support Team.
   
4. The final component in the development group was Lizotte Consultants who has a specific role in meeting the benchmarks and milestones identified in the Work Plan.

Project Lessons Learned

The single greatest lesson learned from this project is that research and development is an unpredictable and risky venture that must be effectively financed to handle the failures that will be encountered or the expected successes will never be achieved.

The areas that were expected to cause issues did but many parameters that were taken for granted caused great difficulty. As expected, Lizotte Consultants had to explore many developments in order for their technology to effectively grade mussels at high speed. In fact, it took three different prototypes to achieve that performance and these units will need more work in the future to reach their full potential.

However, the greatest technical challenge was and still is the application of the Auxiliary Handing Equipment. Thousands of mussels cascading down conveyor belts is a great experiment in unpredictability. Mussels, similar to water, tend to move in many directions and are very difficult to position in a consistent manner at high speed.

However, digital imaging technology is dependent on the object being in a consistent location and format to give a repeatable process. There was a great deal of energy and trial and error spent on designing mussel handling equipment that could effectively singulate and organize mussels in a consistent pattern at high speed. That challenge will continue for years with this new equipment before the 6th or 7th design provides high end results. The handling systems in the actual finished equipment are quite different than those shown in the original design.

3.0 Project Benefits

The benefits of this new technology to the Canadian mussel industry are significant. The industry has been somewhat stagnant in the introduction of automated equipment into the mussel processing component. This introduction of new technology creates the type of paradigm shift in process improvement that can move an industry forward in leaps and bounds. This new automated grader incorporating machine vision technology will change the fundamentals of processing and deliver significant benefits to the industry.

The direct and measurable benefits include lowering and controlling labor costs, improving both quality and food safety performance and making the whole grading process more consistent and predictable.

However, there are also a great deal of indirect benefits that occur from a project of this nature. They include:

• Quality Improvement                                                                                      

  Mussel processors will be able to meet more stringent quality assurance specifications.

• Food Safety Improvement   

  The new mussel grading unit will improve food safety by ensuring that all cracked and/or crushed mussels are identified and removed.

• Technological Advancement

  The processing industry has been stagnant in the introduction of new technology over the last decade and this project will change that.

• Technology Synergy

  This introduction of new technology demonstrates instant improvement and has a synergistic effect on processors to identify and create more innovation.
  

• Marketplace Competitiveness

  The application of a new mussel grading line will provide a new level of competitiveness for the industry.

• Productivity Improvement

  Mussel processors will gain instant productivity improvement through a more effective and dependable grading process.

4.0 Project Deliverables

A key measurement of any project such as this is the determination of its deliverables to the mussel processing industry. This project will have an immediate impact and the industry should experience immediate benefits.
The deliverables are many and include:

1. Availability of an off the shelf, commercial mussel grader using digital imaging technology to grade mussels at a high speed and high level of accuracy.
2. Resulting improved level of global competitiveness for Canadian mussel processors.
3. Canadian producer of manufacturing equipment with a greater capacity to meet the needs of the seafood processing sector.
4. An influx of new technology into the mussel processing sector that will have a synergistic effect on the industry.
5. Less of a dependence on un-skilled labor to perform routine quality control activities.
6. Ability for mussel processors to grade mussels to a much higher level of accuracy and consistency.

5.0 Project Conclusions and Recommendations

There are also a number of conclusions and recommendations that should be articulated for the benefit of future research and the analysis of research assistance programs. They include:

1. Research funding such as the AIMAP of DFO is critical to catalyzing and enhancing and the type of missionary research that must occur to keep an industry globally competitive. It is important that these types of programs receive recognition and funding going forward.
2. It is imperative that the regional development programs such as ACOA and NRC recognize and support these types of innovative research projects. Without the ability to leverage with other funding agencies, these projects would never happen.
3. It would be very easy to take on a project such as this with inadequate funding due the unpredictable nature of technology and the numbers of failures required to achieve success. There would be potential benefits involved in making financial counseling on research and development along with case studies on research projects available to the applicant during the project design stage.
4. The funding models used in project design do not always give enough credit for contributions in kind. This aspect was evident in this specific project. As research challenges were raised in making the technology work, Confederation Cove Mussels Ltd contributed a significant degree more than credited in the budget in both direct costs and those indirect overheads that still have to be paid.
5. A successful project such as this should act as motivation to the industry by indicating that the introduction of new technology is the best approach to move the industry towards a profitable and sustainable future.
6. The research project was successful and the application of this technology worked. The unit currently grades mussels at approximately 4,000 per hour at a high level of accuracy with the potential to get much better. The project has introduced the technology needed for this equipment and the groundwork has been laid for it to get much better in the future.
7. Confederation Cove Mussels Ltd found the Aquaculture Innovation and Market Access Program and staff at the Department of Fisheries and Oceans to be very supportive and effective in making the program work and would be motivated to pursue ventures of this kind in the future.