Did you know? Four classmates can each finish their assigned work on time and still miss the deadline for a group project. The problem may not be laziness or effort. It may be the order of the tasks. A pupil cannot analyse survey answers before the questions have been approved and the survey completed, however many people are available to help.
A task dependency means that one piece of work must be completed, or reach a defined stage, before another can start. To plan a school project realistically, write down the tasks, identify which ones depend on earlier tasks, estimate how long each takes, and calculate the longest necessary chain. That chain—not the sum of every person’s hours—sets the earliest plausible finishing time when independent tasks can proceed in parallel.
This guide shows students, teachers and parents how to make a small dependency map, test a fully worked fictional classroom schedule, diagnose an optimistic plan, practise from guided examples to an independent challenge, and check whether the project is genuinely ready to submit. The aim is a transferable habit: plan the work in the order reality allows, then leave time to check it.
Why a neat task list is not yet a workable plan
Imagine a teacher asking a group to prepare a short display about pupils’ journeys to school. The group writes four tasks on a sheet: make questions, collect responses, design a poster and assemble the final display. Next to each task they write a duration. That list is useful, but it does not tell them which work can overlap. A person may be free to design the poster while another drafts questions; nobody can calculate the final survey totals before the survey exists.
The first diagnostic question is therefore not ‘Who is doing the most?’ It is ‘What must already be true before this task can begin?’ A predecessor is an earlier task whose result is needed. A successor is a task that uses the result. Drawing an arrow from predecessor to successor makes the hidden order visible. The arrows describe logical needs; they do not automatically mean a particular pupil must work on both tasks.
Keep three constraints separate: dependencies concern order; capacity concerns how many people or devices can work simultaneously; deadlines concern the latest acceptable finish. Mixing them can produce a timetable that looks full but cannot actually be executed.
A complete example: can the class display be ready in two and a half hours?
Here is an invented exercise, not a report of an actual eduKate pupil. Two students can work at the same time. All materials and permissions are assumed available at 2:00 pm. There are no breaks, delays or competing lessons in this initial simplified model. The teacher wants the final display ready at 4:30 pm, giving a window of 150 minutes.
- Task A — Draft and check the survey questions: 45 minutes. No earlier task is required.
- Task B — Collect and tally the responses: 60 minutes. It may start only after A.
- Task C — Prepare a reusable poster layout: 75 minutes. It may start immediately; the actual survey numbers are added during final assembly.
- Task D — Insert the totals, check the wording and assemble the display: 35 minutes. Both B and C must be finished first.
Assign Student One to A and then B. Assign Student Two to C. Both help check D. At 2:00 pm, A and C start together. A finishes at 2:45 pm, so B runs from 2:45 to 3:45 pm. C finishes at 3:15 pm. D must wait for both B and C and therefore starts at 3:45 pm, finishing at 4:20 pm. The display appears achievable, with 10 minutes remaining before 4:30.
Hand-check the calculation. The A-to-B chain takes 45 + 60 = 105 minutes. The independent C chain takes 75 minutes. D begins after whichever chain finishes later: max(105, 75) = 105 minutes. Add D’s 35 minutes to obtain 140 minutes. Starting at 2:00 pm, 140 minutes means 4:20 pm. The sum 45 + 60 + 75 + 35 = 215 minutes is the total task effort, not the elapsed time, because two students work in parallel.
The critical chain is the one that controls the finish
In this simplified dependency map, the A → B → D route consumes 45 + 60 + 35 = 140 minutes. The C → D route consumes 75 + 35 = 110 minutes. The first is the longest necessary sequence, often called a critical path when the project is modelled as a task network. C finishes 30 minutes before D can begin, so a delay to C of less than or equal to 30 minutes would not change the planned start of D, assuming no other resource conflict.
The surprise is that helping the shortest task may not make the project finish earlier. If Student Two reduces C from 75 minutes to 50, D still waits until B ends at 3:45 pm; the overall finish stays 4:20. By contrast, if a good design reduces B from 60 to 50 minutes while keeping its output equally reliable, D can start at 3:35 pm and the project could finish at 4:10. This is a scheduling illustration, not a promise that a faster survey is appropriate or ethical.
A changed-condition check: what happens if the survey runs late?
Suppose B actually takes 80 minutes rather than 60. Leave all other durations unchanged. B now runs from 2:45 to 4:05 pm. D therefore runs 4:05–4:40 pm. The group misses the original 4:30 pm deadline by 10 minutes. The delay to B was 20 minutes; the initial finish had only a 10-minute buffer. That is why a plan that finishes at 4:20 is possible under the assumptions, not robust against a modest delay.
A useful response is to agree on a revised scope with the teacher, seek an earlier start if permitted, or rearrange work without weakening its accuracy. Do not quietly invent survey answers, trim the checking stage to zero or claim the display was complete before the data arrived. If the deadline is fixed, the group should explain the constraint and propose a feasible alternative. The more honest the dependency map, the easier that conversation becomes.
How to draw the dependency map with pencil and paper
- Write each deliverable as a verb and an observable output: ‘draft six questions’ is clearer than ‘work on research’.
- Estimate a duration in minutes or hours, including any checking required to call the output finished.
- Draw arrows only where a later task genuinely requires an earlier result. Avoid arrows added merely because a pupil prefers doing things in a particular order.
- Identify tasks that can start together, then calculate the earliest finish for each task from its predecessors.
- Find the latest finishing predecessor at each junction; that controls when the next task can start.
- Compare the predicted finish with the deadline and label the available buffer.
- Add real constraints: one shared computer, limited teacher feedback, safety approval, travel between rooms or a school break.
- Confirm who owns each handover, and record what counts as accepted output.
A simple worksheet can use four columns: task, duration, ‘needs this first’ and earliest finish. But the handover condition deserves its own sentence. ‘Survey collected’ could mean ten unverified responses; ‘survey tally complete’ means totals have been checked and can be used without guessing. A clear definition of done reduces invisible rework.
When the mathematics is right but the plan is still wrong
Mistake one: treating total person-hours as elapsed hours
If two pupils work on independent tasks during the same 30 minutes, they may contribute 60 person-minutes of effort while only 30 minutes pass on the clock. Conversely, two pupils cannot necessarily halve a one-hour task. Some tasks require a single sequence, and coordinating helpers takes time. Always ask whether work is actually divisible.
Mistake two: assuming two tasks can overlap when they use the same resource
Our example assumes two students and enough materials to work independently. If Tasks B and C both require one shared computer for their entire durations, they may have to run sequentially even though their outputs are logically independent. Resource limits can make the true finish later than the dependency-only calculation. The related eduKate guide on planning group tasks around shared equipment explores this difference.
Mistake three: hiding an approval step
If the teacher must check a survey before it is distributed, that approval is a real predecessor. Include the expected turnaround or, if unknown, mark the step as a scheduling risk rather than treating approval as instantaneous. For a sensitive survey, consent, privacy and school rules may matter more than saving minutes.
Mistake four: treating ‘finished’ as ‘submitted correctly’
Saving a poster file and submitting the correct final version are not identical tasks. Before the deadline, verify its title, files, visible labels and submission destination. See how to verify the final version of a group project for a complementary handover routine.
A learner diagnostic: can you see the order without calculating?
Before doing arithmetic, show a student these statements: ‘We can photograph the finished model only after it is built’; ‘We can write the introduction while materials are collected’; ‘We need the teacher’s safety approval before the practical.’ Ask them to draw just three arrows, explain one non-arrow and identify a task that may begin immediately. If they connect everything into one line, they may not yet recognise parallel work. If they connect nothing, they may overlook prerequisites.
For a pupil who struggles, use coloured cards rather than formulas. Place ‘write questions’ left of ‘collect answers’; place ‘poster layout’ on a separate row; bring the two rows together at ‘assemble display’. Only then add minutes. For a confident learner, introduce a resource restriction or a delayed predecessor and ask them to recompute the finish independently.
Guided practice, then independent transfer
Level one — one chain
A student must sketch a diagram (20 minutes), label it (15 minutes) and check the dimensions (10 minutes), in that order. What is the earliest finish? 45 minutes. Have the pupil write the predecessor of each stage and confirm that 20 + 15 + 10 represents elapsed time because none of these tasks overlaps in the stated plan.
Level two — two branches
A presentation requires research (40 minutes) and slide styling (25 minutes) starting together; the script review (15 minutes) requires both. With two available students and no shared-resource problem, the earliest completion is max(40, 25) + 15 = 55 minutes, not 80 minutes. Ask which delay would matter: ten extra minutes of styling would not change the 55-minute finish, but ten extra minutes of research would raise it to 65.
Level three — independent planning challenge
Now try this without looking at a model: select photographs (25 minutes) and gather permission forms (50 minutes) independently. Captions (20 minutes) depend on the photographs; final publication check (15 minutes) depends on captions and forms. Earliest finish is max(25 + 20, 50) + 15 = 65 minutes. The form collection, not the photograph route, controls the finish. Ask the learner to explain why this remains true even though the photo-and-caption route contains two tasks.
For a genuine project, let the student create the map and later compare the estimate with actual start and finish times. A reasonable improvement is not always a shorter forecast. Sometimes the most valuable progress is recognising a previously hidden approval or checking stage before it derails the work.
What parents and teachers can observe
Watch for three behaviours: the pupil names a necessary predecessor without a prompt, identifies what can happen at the same time, and revises the schedule when a duration changes. A brief project diary can record predicted finish, actual finish, main delay and what will be done differently next time. Do not judge success solely by how attractive the finished display looks. The same skills transfer to revision schedules, family plans and later workplace projects.
Teachers can ask learners to justify why two tasks overlap; parents can ask ‘What are you waiting for?’ rather than ‘Have you finished yet?’ The answer helps distinguish a true blockage from an avoidable delay. A calm, specific question often gives the learner more control than a generic reminder to work faster.
Frequently asked questions
Is the critical path always the task with the longest duration?
No. It is the longest required sequence from start to finish, including handovers. Several short tasks in a row can control the finish more than one long independent task.
What if a task can begin when an earlier one is only half-finished?
Split the earlier task into clearly defined stages. For example, ‘first ten survey responses checked’ may allow an initial chart mock-up, while the final chart must wait for all verified data. Never pretend a partly finished output is a complete one.
Is a ten-minute buffer enough?
It depends on uncertainty, task stakes, availability and the deadline’s flexibility. Our fictional ten-minute buffer is small relative to a 140-minute project. It is a calculation, not an official planning recommendation.
Should students calculate all schedules with software?
Not at first. A pencil map and correct reasoning are usually more useful for a small project. A spreadsheet or scheduling application becomes helpful when there are many tasks, changing assumptions and shared resources; the underlying logic still matters.
The deeper learning aim
Education should help students see the structure beneath a busy checklist. The same diagnosis-first principle appears in eduKateSG’s Secondary 1 Mathematics tutorial reference: understand the relationship, practise progressively, then check independently. Explore the Education Hub and the Core Aim of Education learning hub for further applications.
The practical takeaway: Before promising that a group project will be ready, ask which tasks depend on which others. Then let the arithmetic reveal the plan you can actually carry out.