The scenario
A monthly demand and replenishment decision currently made from last month's actuals plus judgement, with no view of the error the current method carries.
Decision AI demo
A forecast attached to a decision someone actually makes
Backtested forecasting and scenario simulation on your history, framed around one recurring decision, with the drivers, the error bands and the recommended action visible together.
In Saudi Arabia
What turns the Decision AI demo into a production system
A working demo and a system a Saudi reviewer will sign off are two different things. These are the questions asked before go-live, and what we hand over against each one.
AuthorityWhat it requiresWhat we buildWhat you keep
SDAIA, PDPL
Lawful basis for every personal-data field the demo touches, before it runs on real records.
Field-level data map produced during scoping, sanitised extract used until that map is signed off.
Data map and lawful-basis note the client's DPO can file before production access is granted.
NDMO
Data classification and quality controls on the sources the demo would depend on in production.
Source-by-source classification, with quality thresholds agreed before the pilot widens.
Classification map and a quality gate report your data office keeps.
NCA ECC
Access control, logging and read/write boundaries clear before wider rollout.
Entitlements scoped per role, every action logged, write paths gated behind approval.
Control matrix and an action log your risk function can replay.
Where it runs
The demo itself can run on a sanitised or representative extract, but where the data cannot leave a system, the build happens inside your environment inside Saudi Arabia, not ours.
Arabic, as it is spoken here
Arabic questions and documents are handled as first-class input, not translated to English before processing, including organisational and Gulf-dialect terms.
Who it is for
Planning, supply chain, finance and revenue owners who commit to a number every month.
Sample workflow
The chain of work behind the answer
1
Frame the decision
- Who decides, how often, on what number
- The cost of being wrong in each direction
- The current method as the baseline
2
Build features
- History, seasonality, Hijri calendar effects
- Promotions, price and external drivers
- Leakage checks before training
3
Backtest
- Rolling-origin evaluation on real history
- Compared against the current method
- Error reported by segment, not just overall
4
Simulate
- Scenario levers the business controls
- Ranges, not single-point certainty
- Driver contribution shown
5
Deliver the decision
- Recommended action with its confidence
- Delivered into the tool people already use
- Outcome tracked after the fact
A model only counts once it beats the current method on your own history, that comparison is the demo, not the chart.
What you will see
Two examples, start to finish
“How much of this SKU should we hold in each region next month?”
- Behind it
- Rolling backtest against last year → seasonality and Hijri calendar effects separated → service-level target applied → cost of overstock vs stockout weighed.
- What comes back
- A quantity per region with an error band, the drivers behind it, and how the recommendation would have performed over the past twelve months.
“What happens to the plan if the price rises five percent?”
- Behind it
- Scenario lever applied → elasticity estimated from history where the evidence supports it → uncertainty widened where it does not.
- What comes back
- A range for volume and margin, with an explicit note where the history is too thin to support a confident answer.
Deliverables
What you keep after the session
Every item below is an artefact your team can reuse, whether or not the engagement continues.
Backtest report
Model versus your current method on your own history, by segment.
Working scenario tool
The levers your planners can move, with ranges rather than false precision.
Decision brief
One page: the decision, the recommendation, the risk and what would change it.
Data readiness findings
What limited the model, and what to fix to lift accuracy next.
Session agenda
- The decision framed before any model is shown
- Backtest walkthrough, including where the model loses
- Two scenarios your planners choose on the day
- What it takes to run this monthly in production
What we need from you
- Two to three years of transactional history
- The current forecast or plan, so there is a baseline
- Known events: promotions, outages, calendar effects
Where data cannot leave a system, the demo is built inside your environment instead of ours.
FAQ
Frequently asked questions
What if our data is messy?
That is the normal starting point, and the backtest makes the cost of it visible instead of hiding it inside an average.
Will you promise an accuracy figure up front?
No. Accuracy is a property of your data and your decision, and we report it measured rather than promised.
Where does the output land?
In the planning tool or report your team already opens, a model that needs a new habit does not get used.
Topic cluster
Explore the topic cluster
Curated links across our Saudi hubs, sector pages and long-form blog research.