Valeur.Products · Time and Cost Exposure
Time and Cost Exposure: what a decision will cost you in days and in money, before you make it.
It shows how much schedule and cost risk a contract carries today, where that risk comes from, and what would change if you made a change — sold under the name Quantum Foresight. Which part of the answer comes from your own record and which part from an assumption is written out separately every time.
This page carries no sample output and no money figure. Every number produced comes out of one contract’s own data; a number printed here would be one we invented for a contract that does not exist. The only way to see your figure is to look at your own contract.
- How many days and how much cost is this contract exposed to today?
- Where does the band sit between the best and the worst outcome?
- If I move the schedule or change the method, how many days do I gain?
- Which costs more in the end, settling or going to arbitration?
It writes what it knows apart from what it assumes
The most fragile part of a forecast is an assumption you cannot see. Here every input is placed in one of three classes before it is used, and only the first two reach the headline figure. Assumptions are not discarded; they sit in their own list with the value they were given.
- Your own record
- The contract’s or the project’s own data: country, value, schedule window. Enters the headline.
- Verifiable source
- Data from outside that has been verified: real weather history, recorded event counts, a risk study logged earlier. Enters the headline.
- Assumption
- A value placed because no data was there. It is marked, it is listed in the output with its value, and it never enters the headline total.
It draws on six separate sources
Schedule and cost risk does not come from one cause, so it is not read from one place. Six sources are called in turn; none of them writes anything to the record, all of them only read.
- Weather — the schedule effect for the window the work falls in, from real archive history. Gives days.
- Loss of output — the productivity drop that comes from disruption, measured against real production records. Gives days and cost.
- Cash — the cost effect from the recorded cash projection. Gives an amount.
- Risk money — the recorded result of a risk study done earlier. It enters the headline only if the real values are on record.
- Jurisdiction — the legal tradition of the country the work is in on delay and liquidated damages. Not a number: the context and the name of the source it rests on.
- Geopolitics — recorded event counts for the project country, its surroundings and the supply line. The three are reported separately; what comes back is the count itself.
ScopeThe last two sources add nothing to the headline figure, but not because they were excluded. Both are of the class that can enter it; they produce no day or amount to add, because there is no honest coefficient that turns a country’s legal tradition into days. Rather than invent one, we hand the finding over as it is. Being impossible to quantify does not stop a finding from being a finding.
It does not fill the gap, it names it
A forecasting tool that meets missing data has two easy options: come back empty, or fill the gap quietly. Both are closed here. Whatever is missing is written down by name and the run carries on with the rest.
If the contract record is not found
If no record matches the id you gave, the sources that came back empty are written down by name, and the six sources are then called anyway. Nothing is invented in their place, and the run does not stop.
If one source fails
If one of the sources fails, it goes into the missing list along with what the failure was, and the turn passes to the next one. Real data from five sources is not thrown away because of the sixth.
See the outcome before you make the change
A forecast is easy to read and close. Here you can work on top of it: you ask for two kinds of change and see how many days each one moves against today’s position. The difference is not estimated; the weather history is pulled again for the new conditions and the calculation is redone.
-
01
Reschedule — same work, new window
You move the same work into a new schedule window. The weather history is genuinely pulled again for that window and the difference is given against today’s position. You supply the dates; without a new start and end the calculation does not run.
RuleWhat you are doing is asking, not deciding. The calculation runs in memory only and nothing is written to the record — no number you see here has changed the state of your project.
-
02
Change of method — same window, new method
You hold the schedule fixed and change how the work is done. The weather history is pulled again for the new method and the difference is again given against today’s position. You choose the method too; without a choice the calculation does not run.
RuleEvery calculation has to find at least two real years of weather history. If it cannot, that line is marked “no data” and left empty; the rest of the board keeps working. If the contract has no verified weather history at all, the board does not open and writes down why — it does not produce half a forecast.
Frequently asked
Is an assumption that does not enter the headline thrown away?
No, nothing is thrown away. Every input marked as an assumption sits in the output in its own list with the value it was given; it is simply not added to the headline total. The output also carries a counter: it records how many inputs fall into each of the three classes, and writes zero for a class that was never used. So you do not have to count the list to see how many inputs the headline is made of and how many were held out of it. That is why an assumption is placed beside the figure rather than hidden: a hidden assumption makes a forecast impossible to audit.
If two sources give no number, why are they in the output at all?
Because a non-numeric finding is still a finding. The jurisdiction source gives the country’s legal tradition on delay and liquidated damages together with the name of the source it rests on; the geopolitical source gives event counts separately for the project country, its surroundings and the supply line. Neither is converted into days or money, because there is no honest basis for that conversion. Keeping what can be quantified and what cannot in the same output but in separate places is better than hiding the second and better than converting it with an invented coefficient.
Does the geopolitical source read news in real time?
No. Instead of calling a live service it reads from event counts recorded earlier. That has two consequences. The good one: it does not depend on an outside service’s rate limit or its downtime, and it keeps working while that service is down. The hard one: if the record has not been filled yet, it writes down how many countries are covered, goes into the missing list, and names the ones it left out. The counts are given raw, without weighting — weighting would be an interpretation, and that interpretation is yours.
If I like a result on the board, has my project plan changed?
No. Every calculation on the board is a question: it runs in memory only and nothing is written to the persistent record. If the contract has no verified weather history — if the site coordinate is not real, for instance — the board does not open at all and writes down why; it does not produce half a forecast. Every difference on the board rests on a real recomputation, not on an assumption. Changing the plan is a separate decision, and it is yours to make.
Let us look at your own contract’s exposure
The only way to see what this page describes would return on your own project is to look at your own contract. Let us walk through an example together: in the first session you will see which sources are full, which are missing, and what the headline would rest on.