Project Selection and Planning
Open in the Padhaiko app1.1 Selection of minor project based on theory/practical subject studied or as per the research or industry needs
Minor Project
A minor project is a small-scale, group-based engineering project carried out under a supervisor. The group (usually 3-4 members) takes a real problem in computer engineering, designs and builds a working system (software, hardware or a hybrid) and then tests, documents and presents it. It is research-based (study, compare or improve a method) or product-based (build a usable system). The course is only 1 credit, so the project must be small enough to finish in one semester.
The topic can come from three main sources (courses you have studied, research needs, or industry and community needs). Choose the source first, then the idea.
| Source | What it means | Example |
|---|---|---|
| Theory subject | Apply a concept studied in class | Use a DBMS course to build a library system with a normalised schema |
| Practical subject | Extend a lab work into a complete system | Turn a microcontroller lab into a sensor-based irrigation controller |
| Research need | Reproduce or improve a published method on a small dataset | Compare two classifiers for Nepali handwritten digits |
| Industry or community need | Solve a local problem seen on a field visit | Billing and stock app for a nearby pharmacy |
Areas the syllabus encourages
Pick a topic from one of these areas: artificial intelligence and machine learning, image processing and computer vision, networks and cybersecurity, audio and natural language processing, electronic devices and embedded systems, big data technologies, Internet of Things (IoT), and web and mobile application development. Where it helps, make a field visit or collect a real-world dataset.
What makes a good minor project topic
| Criterion | Question to ask |
|---|---|
| Relevance | Does it use what we studied in computer engineering? |
| Feasibility | Can 3-4 students finish it in one semester with the tools and time we have? |
| Availability | Do we have the data, hardware, software and a supervisor who knows the area? |
| Novelty or usefulness | Does it solve a real problem, or add something to an existing solution? |
| Learning value | Will every member design, implement and test something? |
Choosing between three candidate topics with a scoring matrix
A group of four students has three ideas: (A) plant-disease detection from leaf photos with a CNN, (B) a smart-home IoT controller, (C) a network intrusion detection system. Each is scored from 1 (poor) to 5 (excellent) on four criteria with weights: Relevance 0.3, Feasibility 0.3, Availability of data and hardware 0.2, Novelty 0.2.
Scores (Relevance, Feasibility, Availability, Novelty): A = (4, 4, 3, 4), B = (3, 5, 5, 3), C = (5, 2, 2, 5). Which topic should the group choose?
Weighted score of A (plant disease): Multiply each score by its weight and add: 0.3 x 4 + 0.3 x 4 + 0.2 x 3 + 0.2 x 4 = 1.2 + 1.2 + 0.6 + 0.8 = 3.8.
Weighted score of B (smart home): 0.3 x 3 + 0.3 x 5 + 0.2 x 5 + 0.2 x 3 = 0.9 + 1.5 + 1.0 + 0.6 = 4.0.
Weighted score of C (intrusion detection): 0.3 x 5 + 0.3 x 2 + 0.2 x 2 + 0.2 x 5 = 1.5 + 0.6 + 0.4 + 1.0 = 3.5.
B (smart-home IoT) scores highest with 4.0, then A with 3.8, then C with 3.5. The group should choose B, or discuss A if the members prefer machine learning.
1.2 Problem identification and scope definition
Problem Statement and Scope
A problem statement is a short description (3-5 sentences) of who has what problem, why it matters and what is wrong with the present solution. The scope states exactly what the project will do (in scope) and will not do (out of scope), so that the work can be finished and judged.
1.2.1 Identifying the problem
Good problems are found by observation, not by choosing a technology first. Use these methods:
- Observe and interview users (a field visit to an office, shop, school or farm) and write down what is slow, manual or error-prone.
- Review existing systems and note their gaps (cost, language, accuracy, offline use).
- Study the courses you have done and ask which problem they could solve.
- Check data and hardware you can really obtain before fixing the problem.
1.2.2 Defining objectives and scope
State the general objective (one sentence) and 3-4 specific objectives that are SMART: Specific, Measurable, Achievable, Relevant, Time-bound.
| Vague | SMART |
|---|---|
| Make a good attendance system | Build a face-recognition attendance system that identifies at least 90 percent of enrolled students in a classroom test within 3 seconds per student, in 12 weeks |
| Improve the network | Monitor 10 devices on a LAN and raise an alert within 1 minute when a device goes offline |
| Part | Content |
|---|---|
| In scope | Features, users, platform, dataset size, hardware that the project will deliver |
| Out of scope | Features postponed or not attempted (for example multi-language support) |
| Assumptions | Things taken as true (for example stable Wi-Fi in the lab) |
| Constraints | Time (one semester), budget, equipment, team size |
| Deliverables | Working system, source code, report, presentation |
Writing a problem statement and scope for a pharmacy stock app
During a field visit a group finds that a small pharmacy records stock in a notebook, often runs out of fast-moving medicines and sells expired ones. Write a problem statement, a general and specific objectives, and the scope.
Problem statement (who, what, why, present solution): The pharmacy staff track about 400 medicines in a paper register. Searching and counting take long, expiry dates are missed and shortages are noticed only when a customer asks. A paper register gives no alerts and no sales summary.
Objectives: General objective: develop a mobile and web application that manages medicine stock and warns about low stock and expiry. Specific objectives: (1) store medicine, batch and expiry data in a database; (2) alert when stock falls below a set level or when expiry is within 30 days; (3) generate a daily sales and stock report; (4) test with real data from the pharmacy.
Scope: In scope: stock entry, billing, low-stock and expiry alerts, daily report, login for 2 staff roles. Out of scope: online ordering, payment gateways, supplier integration. Assumption: the pharmacy has a smartphone and internet access. Constraint: one semester, a team of four.
The problem statement names the user (pharmacy staff), the problem (manual register, expiry and shortage), and its impact. The objectives are measurable and the scope separates what will be built from what will not.
1.3 Approval of project by the department
Project Approval
Approval is the formal acceptance of the group, the supervisor, and the project topic by the department (usually a project committee). Work should begin only after approval, because the committee checks that the topic is suitable, not duplicated, and feasible.
1.3.1 How the approval process usually runs
The exact rules are set by the department, so always follow its notice. A typical sequence is:
| Stage | What the group does | What the department does |
|---|---|---|
| Group formation | Form a group of 3-4 members and list names and roll numbers | Allots a supervisor |
| Topic discussion | Meet the supervisor with 2-3 candidate topics | Supervisor guides and narrows the choice |
| Submission | Submit the title, problem statement, objectives and a short abstract | Committee reviews the submission |
| Decision | Revise as advised | Approves, approves with changes, or rejects |
| Start of work | Begin planning and development | Monitors through progress meetings |
1.3.2 What to prepare for approval
- Project title that states the system and its purpose.
- Problem statement and specific objectives (from 1.2).
- Scope and expected outcome.
- Names of members and the supervisor, with each member's role.
- Tools and resources needed (hardware, datasets, software) and who provides them.
- A rough timeline for the semester.
Checking that the plan fits the available hours before submission
A group of 4 students can each give about 6 hours per week for 14 weeks. Their estimate of effort is: requirements and study 30 h, design 60 h, development 90 h, testing 70 h, documentation 40 h, presentation 30 h. Is the plan feasible enough to submit for approval?
Hours available: 4 members x 6 hours x 14 weeks = 336 hours.
Hours required: 30 + 60 + 90 + 70 + 40 + 30 = 320 hours.
Compare: 336 - 320 = 16 hours of spare time, only about 5 percent.
The plan needs 320 h against 336 h available, leaving 16 h. It fits on paper but has too little buffer, so the group should trim the scope before seeking approval.
Practice Exercises: Project Selection and Planning
Define a minor project. List the main areas of computer engineering from which a topic may be selected.
Definition: A minor project is a small, group-based project (3-4 students) carried out under a supervisor in which the group identifies a problem, designs, implements and tests a system, and documents and presents it in one semester.
Areas: Artificial intelligence and machine learning, image processing and computer vision, networks and cybersecurity, audio and natural language processing, electronic devices and embedded systems, big data technologies, IoT, and web and mobile application development.
A minor project is a one-semester, supervised, small-scale group project in computer engineering, with a topic from areas such as AI/ML, vision, networks and security, NLP, embedded systems, big data, IoT or web and mobile development.
Explain the three possible sources from which a minor project topic can be selected. Give one example of each.
Theory or practical subjects: Apply or extend something studied in class, for example a hospital database system from the DBMS course.
Research needs: Reproduce or improve a published method on a small dataset, for example comparing two classifiers for spam detection.
Industry needs: Solve a real problem found by a field visit, for example a stock and billing app for a local shop.
Topics come from (1) theory or practical subjects, (2) research needs and (3) industry or community needs, with examples as above.
Two topics are scored (1 to 5) on Relevance, Feasibility and Data availability with weights 0.4, 0.35 and 0.25. Topic P = (5, 3, 4), Topic Q = (3, 5, 4). Which has the higher weighted score? What else should the group consider?
Topic P: 0.4 x 5 + 0.35 x 3 + 0.25 x 4 = 2.0 + 1.05 + 1.0 = 4.05.
Topic Q: 0.4 x 3 + 0.35 x 5 + 0.25 x 4 = 1.2 + 1.75 + 1.0 = 3.95.
Judgement: P is slightly higher, but the difference is only 0.10. Team interest, supervisor advice and the risk of the low feasibility score (3) should also be considered.
P scores 4.05 and Q scores 3.95, so P is higher; the gap is small, so feasibility risk and supervisor advice should settle the choice.
What is scope creep? How can a group prevent it? Rewrite the vague objective "make a good library app" as a SMART objective.
Scope creep: Scope creep is the uncontrolled addition of features during the project, which delays the work and lowers quality.
Prevention: Write the in-scope and out-of-scope lists, rank features as must-have and nice-to-have, build must-haves first, and agree changes with the supervisor.
SMART objective: Build a library app that lets students search 5000 book records and reserve a book in under 3 clicks, tested with 20 users by week 12.
Scope creep is uncontrolled growth of features. It is prevented by a written scope, priorities and supervisor-approved changes. A SMART version: "Build a library app to search 5000 records and reserve a book in under 3 clicks, tested by 20 users by week 12."
Why does the department approve a project before work begins? List what a group prepares for approval. A group of 3 plans 5 hours per week each for 12 weeks. If the work estimate is 180 hours and a 20 percent buffer is wanted, how many hours should be planned, and is this plan feasible?
Purpose of approval: The committee checks that the topic is suitable for the level, not duplicated, feasible with the resources, and that a supervisor is assigned.
What to prepare: Title, problem statement and objectives, scope, member and supervisor names, tools and resources, and a rough timeline.
Hours: Available = 3 x 5 x 12 = 180 hours. With a 20 percent buffer the plan needs 180 x 1.2 = 216 hours, more than the 180 available, so the plan is not feasible: reduce the scope or add hours.
Approval ensures suitability, originality, feasibility and supervision. The group submits title, problem, objectives, scope, team, resources and timeline. With a buffer the plan needs 216 h but only 180 h are available, so the scope must be reduced.