RECYCLE RIGHT THE FIRST TIME WITH MACHINE VISION & AI.

Project: Build A Machine
Vision System To Help Sort
Recycled Goods
Recycling is important to reduce our environmental and ecological footprint, as you know. The college is
committed to maximizing the proportion of goods recycled while minimizing the amount of waste destined
for landfill. To this end the college has installed recycling stations throughout the campus. Each recycling
station has several bins, and users are expected to sort their used containers and waste material into the
correct bins. The college pays a private contractor to empty and process these bins.
Unfortunately, despite clear labelling, many users do not correctly sort their used containers and waste
products. This is a problem for the college, since the recycling contract stipulates that material is to be
correctly sorted. The college pays a penalty each time a recycling bin is contaminated with material that
should be in a different bin. The aggregate of these recycling penalties is a significant expense for the
college.
To mitigate this expense, the college is looking for solutions. One solution under consideration is to outfit
each recycling station with a machine vision system and train it to recognize common containers and waste
items. Once it recognizes an item, it lights up an indicator next to the appropriate bin, guiding users to
recycle correctly. In addition, the system sends alerts to facilities personnel so they know when a bin is
approaching capacity.
You have been asked to prepare a project plan to implement this solution, including basic measures of risk
and economic viability.
SCOPE OF WORK
The course project scope includes a mix of predictive and adaptive project management activities:
• Prepare a product vision statement.
• Prepare a project charter.
• Identify and rank key stakeholders in a stakeholder map.
• Identify at least three overall project risks and outline mitigation measures.
• Collect and manage product requirements in a product backlog.
• Adjust the product backlog for risk and prioritize by value.
• Create a sprint backlog.
• Identify at least one architectural or technical spike.
• Prepare information radiators based on simulated development progress.
• Prepare a WBS and project timeline for deploying the system (hint: https://www.teamgantt.com)
• Allocate and level resources in the deployment project timeline.
• Determine the predicted completion date and the BAC.
• Calculate metrics of project economic viability (ROI, NPV, IRR, 5 year horizon, 12% discount rate).
• [Instructor discretion] Present your plan.
Note: this is not a programming assignment. You are not required to do any coding or technical product
development. Your instructor will provide you with simulated updates from your development team.
DETAILS
Assignment Type: Teams of 2 or 3
Original Work: Required — do not copy an online example or the work of others
Submission: Please use the links provided by your instructor
Due Date: There are several due dates for staggered deliverables; dates are assigned in class
Writing Standard: Business professional
Cover Page: Required
Author(s) Name(s): Required
Course Code: Required
APA Style: Ensure any material drawn from a third party is appropriately cited
HOW TO GET TOP MARKS
Remember that uncertainty is part of all projects; work with what you have and reach out if you need
additional information; study the problem and the attached information carefully; consider who your key
stakeholders would be; identify user needs (your instructor will serve as a proxy); create a detailed and
realistic product backlog; progressively elaborate product backlog items into tasks and acceptance criteria
in a sprint backlog; build a detailed deployment plan; consider project risk; calculate realistic measures of
economic viability; avoid submitting incomplete or superficial work; don’t plagiarize; follow submission
guidelines properly and submit on time!
ASSESSMENT CHECKLIST
Build A Machine Vision System
To Help Sort Recycled Goods
Completeness of vision statement & project charter /10
Completeness of stakeholder map & project risk assessment /10
Completeness of product backlog /20
Completeness of sprint backlog & information radiators /20
Completeness and accuracy of predictive deployment plan /20
Completeness and accuracy of economic viability metrics /10
Overall impression of quality and presentation /10
Total /100
Days late (@5)
Spelling/grammar faults (@5)
Plagiarized or unoriginal work (0% + academic sanction)
Total adjusted /100
ACME RECYCLING CORP
333 Riverfront Avenue SE
Calgary Alberta T2M2Q4
CLIENT: Bow Valley College, 345 6 Avenue SE, Calgary, AB T2G 4V1
ACCOUNT NUMBER: 49322-39232
YOUR RECYCLING STATEMENT
January 1 2018 to December 31 2018
Gross recycled tonnage 61234 kg
Organics 12840 kg $28894.50
Plastics & glass 38219 kg $101280.35
Paper & cardboard 10175 kg $22893.75
Cross-contamination penalties $38750.00
Total recycling charges $191818.60
GST $28772.29
Statement total $220591.39
Payment received with thanks ($220591.39)
Account balance $0
To: Dave Johnson, Facilities Manager, Bow Valley College
From: Sarah Williams, Recycling Coordinator, Bow Valley College
Date: March 1, 2019
Hi Dave,
As you requested, I tallied up all the recycling stations across the North
and South campuses at the college. It turns out we have a total of 65
stations.
I also figured out the rates for people at the college who can help with the
machine vision deployment project:
Installer, $200/day, 4 hours to install each system, 3 available
Machine Learning Trainer, $250/day, 16 days to train system, 4 available
Designer, $400/day, 2 days to create posters for stations, 1 available
Communications, $400/day, 1 day to update college website, 1 available
Printer, $0.50 per poster, 1 day turnaround for batch, 1 available
Guide, $300/day, 2 hours at each station to demonstrate system, 2 available
IT Support, $350/day, 2 days to set up networking and messaging, 1 available
The above assumes our usual college work schedule of 7 hours per day and 5
days per week.
I hope this helps!
Sarah
EAI Fabrication Limited
200-1330 72 Avenue SE
Calgary, Alberta
T2M 4W6
David Johnson, Facilities Manager
Bow Valley College
345 6 Avenue SE,
Calgary, Alberta
T2G 4V1
March 30, 2019
Dear David,
With pleasure, I attach herein our quotation for the custom fabrication of machine vision
enclosure units as we discussed. To summarize:
● Custom floor-standing enclosure, painted to match recycling bins
● Forward aiming high-resolution camera with autofocus
● Embedded IOT computer for machine vision and image processing
● Indicator lights for recycling bins
● Delivered to your facility
We can fabricate these units at a cost of $850.00 each plus GST. Please note that software for
the embedded IOT computer is not included in this quotation.
We can deliver units at a rate of 35 per month and will bill you monthly upon delivery.
I trust you find everything in order. Don’t hesitate to get in touch if I can assist further.
Steve Emerson,
Fabrication Estimator
Engaging & Awesome Inc
To: David Johnson, Facilities Manager, Bow Valley College
From: Michelle Chan, Development Manager, Super Software Company
Date: March 21, 2019
Good morning David,
It was a pleasure discussing the machine vision recycling project with you
today. I’m happy to report that this is absolutely within our capability as
we have done similar projects recently.
I think for this project you need an Agile/Scrum development team with 5
members: a product owner, a scrum master, and three full-stack developers.
I can offer you this team at a rate of $550/day/person, which works out to
$2750/day for the team. The team will work on your project exclusively, 5
days per week, until it is completed to your satisfaction. We’ll use twoweek
iterations and will have a product increment to demonstrate at the end
of each sprint.
If this is acceptable to you we can begin identifying stakeholders and
eliciting requirements right away.
I look forward to working with you. Sincerely,
Michelle Chan
Software Engineering Graduate, Bow Valley College (2019)