Lesson 6 of 8

Module 6 · Project Manager Capstone: Rescue the Opening

The forecast

Turn week 10's progress into a forecast of the finish cost and the opening date, using the remaining work rather than a hopeful guess, and give the sponsor a range with a stated basis.

About 35 minutes

Open in ColabEvery example as a notebook you can run, edit and save to your Google Drive.

The problem

You know where the project stands: behind, and over budget. The sponsor's next question is "so what will it finish at?" There are two tempting answers, and both are wrong. One is to say "we'll catch up", which assumes the problems of the past ten weeks will vanish. The other is to extrapolate a single index across the whole project, which assumes everything will go as badly as the worst part has.

A good forecast starts from what's left to do, says how fast it's likely to go, and gives a range. For cost, EVM supplies standard estimates at completion. For the date, you re-run the schedule on the remaining work.

The concept

Forecasting the cost: estimate at completion (EAC)

Three common versions, each with an assumption:

EAC	Formula	Assumes
Typical	BAC ÷ CPI	The rest of the project costs as much per unit of work as so far
Atypical	AC + (BAC − EV)	The overrun so far was a one-off; the remaining work costs what was budgeted
Re-estimate	AC + a new bottom-up estimate of the work remaining	You know more now and have redone the estimate

The estimate to complete is ETC = EAC − AC, and the variance at completion is VAC = BAC − EAC (negative: over budget at the end). When CPI has been stable or falling, the typical EAC is the safer planning figure. The atypical one is the best case.

Forecasting the date: remaining work, re-scheduled

SPI isn't a date. To forecast the finish:

Mark each finished task as done (zero remaining duration).
For tasks in progress, estimate the remaining duration, either at the planned rate (the remaining percentage of the likely duration) or at the rate actually achieved so far (elapsed days ÷ percent done × percent left).
Leave tasks not started at their likely duration.
Run the critical path on what's left, and add the result to today's date (day 50).

The planned-rate forecast is the hopeful end of the range. The current-rate forecast is the realistic one when a task has genuinely been slower than planned.

An invented cost forecast two ways and a date forecast at the planned and the current rate, reported as a range with its basis — Two cost forecasts and two date forecasts, each with its assumption.TapClick the image to see it full size.

Example

Set up, with the schedule function from lesson 3:

Python · try it yourself

import numpy as np
import pandas as pd

base = "https://academy.cloudtechanalytics.com/datasets/lab/"
tasks = pd.read_csv(base + "tasks.csv").fillna({"predecessors": ""})
status = pd.read_csv(base + "weekly_status.csv")
START = np.datetime64("2026-11-02")
PROMISE_DAY = 91

def schedule(tasks, durations):
    preds = {t: [p for p in ps.split(";") if p] for t, ps in zip(tasks["task_id"], tasks["predecessors"])}
    es, ef = {}, {}
    for t in tasks["task_id"]:
        es[t] = max((ef[p] for p in preds[t]), default=0)
        ef[t] = es[t] + durations[t]
    end = max(ef.values())
    ls, lf = {}, {}
    for t in reversed(list(tasks["task_id"])):
        successors = [s for s in tasks["task_id"] if t in preds[s]]
        lf[t] = min((ls[s] for s in successors), default=end)
        ls[t] = lf[t] - durations[t]
    out = pd.DataFrame({"es": es, "ef": ef, "ls": ls, "lf": lf})
    out["float"] = (out["ls"] - out["es"]).round(6)      # rounded: fractional durations leave tiny crumbs
    return out, end

def finish_date(days):
    return np.busday_offset(START, int(np.ceil(days)) - 1, roll="forward")

tasks["budget"] = tasks["likely_days"] * tasks["daily_cost_ngn"]
BAC = tasks["budget"].sum()
now = status[status["week"] == 10].set_index("task_id")
EV = (now["percent_complete"] / 100 * tasks.set_index("task_id").loc[now.index, "budget"]).sum()
AC = now["actual_cost_ngn"].sum()
CPI = EV / AC

Edit the code and press Run (Ctrl + Enter). Examples on this page share their variables, like cells in a notebook.

The cost forecasts first:

Python · try it yourself

Edit the code and press Run (Ctrl + Enter). Examples on this page share their variables, like cells in a notebook.

Output

BAC:                ₦90,750,000
Typical EAC (BAC/CPI): ₦120,337,860   over by ₦29,587,860 (33%)
Atypical EAC:          ₦107,924,000   over by ₦17,174,000 (19%)

Even the hopeful case is ₦17m over budget because the overrun already happened. If the remaining work costs as much per unit as the work so far, the project finishes ₦29.6m over, a third more than the sponsor approved. Neither includes a naira of the changes the sponsor is being asked to consider.

Now the date. Work out each task's remaining duration at the two rates, then reschedule what's left:

Python · try it yourself

first_week = status.groupby("task_id")["week"].min()

def remaining(rate):
    rem = {}
    for t, likely in zip(tasks["task_id"], tasks["likely_days"]):
        if t not in now.index:
            rem[t] = likely                                   # not started
            continue
        pc = now.loc[t, "percent_complete"] / 100
        elapsed = 5 * (10 - first_week[t] + 1)                  # days since it started reporting
        if pc >= 1:
            rem[t] = 0
        elif rate == "plan":
            rem[t] = likely * (1 - pc)                          # the rest at the planned pace
        else:
            rem[t] = elapsed * (1 - pc) / pc                    # the rest at the pace achieved so far
    return rem

for rate in ("plan", "current"):
    left, remaining_days = schedule(tasks, remaining(rate))
    finish_day = 50 + remaining_days
    print(f"At the {rate} rate: {remaining_days:.1f} days of work left, opening on {finish_date(finish_day)}"
          f" ({int(np.ceil(finish_day)) - PROMISE_DAY} working days after the promise)")

Edit the code and press Run (Ctrl + Enter). Examples on this page share their variables, like cells in a notebook.

Output

At the plan rate: 49.2 days of work left, opening on 2027-03-19 (9 working days after the promise)
At the current rate: 54.7 days of work left, opening on 2027-03-26 (14 working days after the promise)

Whichever rate you take, the promised date is gone: the opening is somewhere between Friday 19 March and Friday 26 March, 9 to 14 working days after the promise. The critical chain is unchanged (analyser import, installation, reagents, validation, inspection, trial run), and it starts with the one task that is behind.

Python · try it yourself

Edit the code and press Run (Ctrl + Enter). Examples on this page share their variables, like cells in a notebook.

Output

Critical now: B4 -> B5 -> E2 -> F1 -> F2 -> F3 -> F4

Walkthrough

Set up the schedule function, BAC, EV, AC and CPI as in lesson 5.
Calculate the typical and atypical EAC, and the variance at completion for each.
Calculate each task's remaining duration at the planned rate.
Re-run the schedule on the remaining durations, and add the result to day 50.
Repeat at the current rate.
Compare each forecast with the promised day.
Write the forecast for the sponsor (the task below).

Practice

What is the typical EAC (BAC ÷ CPI), rounded to the nearest naira?

Practice

If the remaining work goes at the planned rate, on what date does the project finish? Type it as YYYY-MM-DD.

Practice

At the current rate, how many working days after the promised day does the project finish?

Task

12 min

Write the forecast for the sponsor (80 to 170 words): the opening date range and its basis, the cost forecast with the assumption behind each version, how far the promised date is missed, and what you'd not want them to assume.

Your work is checked for

Gives a date range (19 to 26 March)
Gives the upper date (26 March)
Says the promised date is missed
Gives a cost forecast (107.9 or 120.3 million)
States an assumption (rate, one-off, CPI)
Says what not to assume (catch up, no changes)
Between 80 and 170 words

Check your understanding

Answer every question to check.

01The problem

02The concept

Forecasting the cost: estimate at completion (EAC)

Forecasting the date: remaining work, re-scheduled

03Example

04Walkthrough

05Practice

06Check your understanding

01The problem

02The concept

Forecasting the cost: estimate at completion (EAC)

Forecasting the date: remaining work, re-scheduled

03Example

04Walkthrough

05Practice

06Check your understanding

The problem

The concept

Example

Walkthrough

Practice

Check your understanding

The problem

The concept

Example

Walkthrough

Practice

Check your understanding