Project Panday

Project Panday: Content Reviewer & Proposal Defense Questions

An A.I-Powered Mobile System for Construction Project Planning and Skilled Worker Recommendation

Part 1: Chapter-by-Chapter Content Reviewer | Part 2: Anticipated Panel Questions

Part 1 — Content Reviewer

Chapter 1 — Introduction

Background of the Study

Construction is a major driver of economic growth but remains one of the least digitalized industries. Digitalization, automation, and Industry 4.0 technologies improve communication, resource coordination, and productivity. Artificial Intelligence supports planning, monitoring, and resource management by analyzing project data.

In the Philippine setting, homeowners still rely heavily on word-of-mouth and personal recommendations to find workers, and Filipino skilled workers often depend on manpower agencies or informal channels. A Metro Manila study found over 89% of respondents struggle to verify a worker's competence and credentials. These gaps — weak planning support and unreliable worker selection — motivate Project Panday.

- Legal Basis: Legal basis: R.A 10173, R.A 8792 (Electronic Commerce Act), R.A 10844 (creation of the Dict).

- S.D.G Alignment: S.D.G alignment: S.D.G 8 (Decent Work and Economic Growth), S.D.G 9 (Industry, Innovation and Infrastructure), S.D.G 11 (Sustainable Cities and Communities).

Statement of Objectives

- To identify the current processes in a construction project.

- To determine the features of the proposed system.

- To test the usability of the proposed system.

Conceptual Framework

The study uses the Input-Process-Output (I.P.O) Model. Inputs are customer needs, project data, skilled worker profiles, and system data, gathered through interviews and surveys with stakeholders. The Process stage applies the Rapid Application Development (R.A.D) Model — requirements planning, user design (prototype cycles of prototyping, testing, and refining), construction, and cutover. The Output is the Project Panday mobile application, offering A.I-based project planning, skilled worker recommendation, and resource estimation.

Scope and Limitations

The system covers building, renovation, repair, and home improvement of residential houses only. Core features include registration/authentication, profile management, project posting, skilled worker profiling, A.I-based recommendation, job matching, and notifications.

- Out of scope: Excluded: commercial/industrial projects, payment processing, contract administration, real-time location tracking, direct supervision, and workforce performance monitoring.

- Important disclaimer: The A.I recommendation feature assists decision-making only — it does not guarantee hiring, work quality, or worker availability.

Significance of the Study

- Contractors: Contractors — easier identification of skilled workers and client connections.

- Homeowners: Homeowners — faster, more informed hiring decisions.

- Skilled Workers: Skilled Workers — greater visibility and job opportunities.

- Future Developers: Future Developers — a reference for enhancing or expanding the system.

- Developers: Developers (Researchers) — practical experience in A.I, mobile development, and recommendation systems.

Key Terms to Know

- Homeowner: Homeowner — property owner who requests construction services and worker recommendations.

- Client: Client — user who plans construction projects and receives material/worker recommendations.

- Contractor: Contractor — uses the system to identify qualified workers and plan projects.

- Skilled Laborer: Skilled Laborer — carpenters, masons, plumbers, electricians, welders, painters recommended by the system.

- Worker Recommendation: Worker Recommendation — the system's process of matching qualified workers to a posted project.

Chapter 2 — Review of Related Literature

The R.R.L is organized around the study's three objectives:

- Current Construction Processes: Construction planning, material estimation (quantity takeoff), and personnel allocation are critical to avoiding delays and cost overruns. Worker selection traditionally relies on referrals and personal experience rather than verified competence.

- System Features (A.I & Recommendation): A.I and recommendation technologies improve planning, decision-making, and resource optimization. Machine learning underlies effective recommender systems. Mata et al. (2024) found system quality and usability drive technology adoption in Philippine construction; supports R.A.D as a rapid, iterative development approach.

- Usability of Systems: Usability evaluation — commonly through the System Usability Scale (Sus), T.A.M, pacmad, or i.s.o 25010/9241 standards — is essential to identifying design issues and ensuring user satisfaction (Park & Zahabi, 2021; Weichbroth & Giedrowicz, 2024; Afif, 2023; Moumane et al., 2016; Ali et al., 2022; Ampuan & Delena, 2022, which reported a 90.2% Sus rating for a comparable local system).

Chapter 3 — Methodology

Research Design

The study uses the Descriptive-Developmental Research Design — descriptive in documenting current construction planning/worker-selection practices, and developmental in designing, building, and evaluating the mobile application itself.

Development Methodology — R.A.D Model

- 1. Analysis and Quick Design — interviews, surveys, and consultations with homeowners, contractors, and skilled workers to define system requirements and initial architecture.

- 2. Prototype Cycle (Build arrow Demonstrate arrow Refine) — an initial working prototype is built, tested with users, and iteratively refined based on feedback.

- 3. Testing — functional and usability testing to verify the system meets operational requirements.

- 4. Implementation — deployment of the validated system for actual use.

Source of Data

- Primary: Primary data: interviews with 1 contractor and 1 general manager; survey questionnaires distributed via Google Forms to homeowners, construction workers, and I.T specialists.

- Secondary: Secondary data: journals, books, published theses, and government publications on construction management, A.I, and mobile app development.

System Architecture

Three-Tier Client-Server Architecture:

- Presentation Layer: Presentation Layer — React Native mobile application (user interface).

- Application Layer: Application Layer — Laravel backend handling authentication, business logic, worker recommendation, cost estimation, and notifications.

- Data Layer: Data Layer — My SQL database storing user accounts, profiles, project data, and recommendations.

Layers communicate through RESTful A.P.I's over H.T.T.P/H.T.T.P.S.

Technology Stack

- Frontend: Frontend: React Native, JavaScript, Expo, React Navigation, Axios.

- Backend: Backend: Laravel, P.H.P, My SQL, phpMyAdmin, xampp.

- Hardware: Minimum dev hardware: Intel i5 (10th Gen)/Ryzen 5, 8 G.B R.A.M (16 G.B recommended), 256 G.B S.S.D; Android 10+ device with 8 G.B R.A.M, 64 G.B storage.

Instrumentation & Data Analysis

- Instruments: Semi-structured interviews — with the contractor and general manager, analyzed using Thematic Analysis.

- Survey questionnaires — 3 sections (demographics, current practices, system evaluation), distributed via Google Forms.

- System Usability Scale (Sus) — 10 standardized statements on a 5-point Likert scale, analyzed using Weighted Mean (W.M = Σfx / N).

Weighted Mean interpretation: 4.21 to 5.00 Strongly Agree | 3.41 to 4.20 Agree | 2.61 to 3.40 Neutral | 1.81 to 2.60 Disagree | 1.00 to 1.80 Strongly Disagree.

Part 2 — Anticipated Panel Questions

These questions are organized by chapter, based on your submitted manuscript. Use them to rehearse concise, confident answers — most panels probe why a choice was made, not just what it is.

Chapter 1 — Introduction

Background of the Study

1. Why did you choose the construction industry specifically, and not another sector that also lacks digitalization?

2. Your background cites many foreign studies (McKinsey, World Economic Forum, etcetera) — how do these apply to the local (Philippine/Dagupan) context?

3. You mentioned Republic Act No. 10173 (Data Privacy Act) — what specific personal data will your system collect, and how will you protect it technically (not just legally)?

4. How does Project Panday differ from existing platforms like handyman/job-matching apps (e.g., the "Handy Fix" study you cited)?

5. You cited a study claiming 89% of respondents in Metro Manila couldn't verify a worker's competence — how will your system actually verify or validate a skilled worker's credentials?

6. What made you decide to focus only on residential construction and exclude commercial/industrial projects?

Statement of Objectives

7. Your objectives are broad ("identify current processes," "determine features," "test usability") — can you state measurable success indicators for each objective?

8. How will you know if Objective 1 (identifying current processes) has been successfully achieved?

9. Objective 3 mentions testing usability — what usability score or benchmark will define "acceptable" for your study?

Conceptual Framework

10. Walk us through your I.P.O model — trace one specific example (e.g., a homeowner's project request) from Input to Output.

11. Why did you choose the R.A.D model over other S.D.L.C models like Agile, Scrum, or Waterfall?

12. In your framework, "skilled workers" appear connected to the Refine/Testing stage — what exactly is their role in that stage versus the homeowners'?

Scope and Limitations

13. You stated the system does not handle payment processing or contract administration — how will actual transactions and agreements between homeowners and workers be handled outside the app?

14. Since the A.I recommendation "does not guarantee work hiring, quality of work, or availability" — what liability protections or disclaimers will be built into the system?

15. Why is the system limited to Android only, and not iOS?

Significance of the Study

16. Among your five beneficiary groups, which one benefits the least directly, and how would you justify including them?

Definition of Terms

17. How do you distinguish "Contractors" from "Small Construction Businesses" operationally within the system (do they have different account types/permissions)?

Chapter 2 — Review of Related Literature

18. Of all the cited studies, which one most directly informed your A.I recommendation algorithm's actual logic/design?

19. You cited both foreign and local (Philippine) literature — what specific gap in the local literature does Project Panday address that foreign studies do not?

20. Several cited works (e.g., Regona et al., Darko et al.) discuss A.I broadly in construction — none seem to describe a recommendation algorithm identical to yours. What recommendation technique (content-based, collaborative filtering, hybrid, rule-based) will you actually implement, and why?

21. You cited studies using Sus across many different systems (appointment systems, L.M.S, information systems) — why is Sus an appropriate fit for a marketplace/recommendation-style mobile app specifically?

22. How did the synthesis at the end of each R.R.L section directly translate into a system feature or requirement?

Chapter 3 — Methodology

Research Design

23. Why Descriptive-Developmental Design specifically, instead of pure Developmental Research or Design Science Research (D.S.R), which is more common for I.T capstones?

R.A.D Model / Phases

24. In the Analysis and Quick Design phase, what specific tools or diagrams (use case, E.R.D, D.F.D) did you produce, and can you show them?

25. During Prototype Cycles (Demonstrate to Refine), how many iterations do you plan, and what changed between iterations?

26. What is your exit criterion for moving from Testing to Implementation — that is, what result would tell you the system is not ready?

Source of Data / Respondents

27. You conducted interviews with only one (1) contractor and one (1) general manager — is this sample sufficient to generalize"current industry practices"? How do you address this limitation?

28. How many total respondents will complete your survey questionnaire and Sus evaluation, and how were they selected (sampling technique)?

29. How do you ensure the skilled workers you sample represent multiple trades (carpentry, masonry, electrical, plumbing, painting) and not just one specialization?

System Architecture

30. Why a Three-Tier Client-Server Architecture instead of a serverless or microservices approach, given this is a matching/recommendation platform that may need to scale?

31. Walk us through what happens, step by step, from a homeowner submitting a project request to receiving a list of recommended workers  trace it through Presentation  Application  Data layer.

32. How will the system remain responsive/available if the internet connection is unstable, given it's described as fully dependent on internet connectivity?

A.I / Recommendation Component

33. What specific A.I or machine learning technique will drive the "worker recommendation" feature (e.g., rule-based matching, cosine similarity, collaborative filtering, a trained classifier)?

34. What data/features will the recommendation engine use to match a worker to a project (skills, location, ratings, availability, price)?

35. How will the system handle "cold start" — that is, recommending workers or materials when there isn't yet enough historical data (new users, new workers with no ratings)?

36. Is your "A.I" a true machine-learning model, or a rule-based/if-then matching system labeled as A.I? Be ready to justify the term precisely.

Hardware / Software Requirements

37. Why Laravel and My SQL specifically, instead of a more mobile-native backend stack (e.g., Firebase, Node.js)?

38. Why React Native instead of Flutter or native Android (Kotlin/Java) development?

Instrumentation and Data Analysis

39. Why did you choose the System Usability Scale (Sus) over other frameworks like T.A.M, pacmad, or i.s.o 25010 that were also discussed in your R.R.L?

40. How will you interpret a Sus score numerically (e.g., what score, out of 100, will you consider "acceptable" for your system) — do you have a target benchmark (e.g., 68 = average, above 80 = excellent)?

41. You will use Thematic Analysis for the interviews — what coding process or software (manual coding, Nvivo, etcetera) will you use to generate themes?

42. How will Weighted Mean results and Thematic Analysis results be triangulated or related to each other in your final evaluation?

General / Cross-Cutting Questions

43. What makes Project Panday different from existing apps like Kamustahan, TaskUs-type gig platforms, or even Facebook Marketplace groups where homeowners already find workers informally?

44. What is your project's most significant technical risk, and how do you plan to mitigate it?

45. If you had more time/resources, what feature would you add next?

46. How will the system verify that a "skilled worker" profile is truthful (e.g., fake certifications, fake ratings)?

47. What happens if there's a dispute between a homeowner and a worker — does the system have any built-in resolution mechanism, or is that explicitly out of scope?

48. How do you plan to onboard skilled workers, many of whom may not be highly tech-literate, onto a mobile app?

49. What is your project timeline for Chapters 4 to 5 (development, testing, actual implementation)?

50. How does your study operationally define and measure "success" of the whole system — is it usability alone, or also actual adoption/matching outcomes?

Tips for the Defense

- Have one clear diagram ready (system architecture, E.R.D, or use case) that you can point to for any technical question.

- For every objective, be ready to name the exact tool/instrument used to measure it (e.g., Objective 3 to Sus).

- Be explicit about what makes your system "A.I-powered" — this is almost always challenged.

- Acknowledge limitations honestly (small interview sample, Android-only, no payment processing) rather than downplaying them — panels respond better to informed awareness than defensiveness.

You have reached the end of the document.