Retail chains

In a distribution network, workload peaks appear when orders and deliveries accumulate in the same receiving and dispatch windows. Everyday disruption adds to the problem: absenteeism and labor shortages, carrier delays, weather-related blockages, or sudden increases and decreases in demand after a campaign or competitor move. Operational flexibility is essential — but it should not come at any cost.

Nadii is designed for these imperfect, real-world conditions. It combines demand forecasts for seasonality, promotions and events with supply forecasts based on probabilistic lead times and supplier service levels such as On Time In Full. It also reflects warehouse throughput, loading and unloading capacity, and carrier constraints.

Based on this, Nadii forecasts node workload in advance and actively shapes flows: moving batches to earlier or later windows, consolidating shipments into full logistics units, adjusting route frequency and size, balancing dispatches between warehouses and — where relevant — coordinating the pace of marketing activity with operational capacity. At the core is one cost-probabilistic decision: when, how much, from where and to where inventory should move. The same decision smooths workload, protects on-time availability and minimizes total cost.

What you gain:

• more stable warehouse throughput and less overtime during peaks,
• lower transport costs through better consolidation and route-frequency planning,
• fewer delays and bottlenecks, with more reliable SLA performance,
• higher availability without simply increasing safety stock,
• one consistent decision model from demand planning through to dispatch scheduling.

Excess stock in one location and shortages in another create a double cost: lost sales on one side and frozen capital on the other. In omnichannel retail, priorities also conflict — e-commerce, marketplaces, stores and B2B have different SLAs, commissions and effects on customer experience and brand reputation.

Nadii looks at the network end to end and treats inventory as a shared resource rather than a set of isolated silos. It can therefore balance stock through relocations before triggering additional purchases, while managing intelligent reservations by channel so online demand does not consume stock needed in stores — and vice versa.

The logic combines consistent demand and supply forecasts with costs. Demand is forecast by channel and location, including promotions, seasonality and events; supply reflects lead times and supplier service levels. Nadii then compares the alternatives: is it faster and more economical to transfer inventory from A to B, or to buy from the supplier? When temporary shortages affect less strategic items, the system estimates potential lost sales and the cost of each option — including margin, customer impact, fees or penalties, transport cost and campaign consequences — then adjusts buffers and reservations to minimize the total business impact.

The balancing scenario is considered already when an order proposal is created. That means fewer unnecessary purchases, more sales from the inventory you already own, and less capital trapped in the wrong place.

What you gain:

• Fewer purchases, more sales — use existing stock first, reducing operating inventory and storage cost.
• Released working capital — excess stock is not left sitting in inactive locations or channels.
• Higher service levels by channel — e-commerce and stores receive inventory when they need it, without cannibalizing one another.
• Fewer markdowns and end-of-line clearances — relocate inventory before resorting to discounting.
• Shorter effective lead times and more stable cash flow — fewer emergency supplier orders and more planned network movements.

One Nadii decision connects demand and supply forecasts, costs, channel SLAs and relocation opportunities. The benefits therefore do not happen independently: the same move can improve availability, avoid a new purchase and release working capital.

Promotions can completely reshape demand: short-lived sales spikes, demand pulled forward from the weeks after a campaign, planogram requirements and mandatory presentation stock. Traditional planning often smooths the history and loses the true campaign effect — leading to empty shelves during the promotion and excess stock after it.

Nadii approaches the problem end to end. It builds daily forecasts for each product in each store and channel, separating the effects of promotion, seasonality and substitution. Those forecasts are combined with the campaign calendar, presentation-stock and planogram requirements, and supply realities such as lead times, supplier service levels and minimum order quantities. Nadii can therefore plan the initial build-up before a campaign and adjust replenishment during the promotion according to actual sales pace, while protecting the required shelf presence.

The system automatically identifies the last safe order date, forecasts the post-campaign stock run-down and — where justified — recommends transfers between stores instead of new purchases. It also supports reservations for individual channels such as stores, e-commerce and B2B so they do not unintentionally consume one another's inventory. Excess stock in one location is considered for relocation before another order is placed.

Every recommendation weighs shortage risk against excess cost while respecting logistical constraints and campaign objectives. Reporting makes the cost of presentation stock visible and highlights cases where it is disproportionate to the sales potential of the store. The result is better availability during the campaign, less leftover stock afterwards, less capital tied up, stronger margin and smoother operations.

What you gain:

• fuller shelves during promotions without carrying unnecessary stock after the campaign,
• less working capital tied up because existing inventory is used before new stock is bought,
• higher service and margin through fewer stockouts at peak demand and fewer emergency markdowns,
• clear visibility into the true cost of presentation stock by store and category,
• smoother operations with fewer workload spikes and last-minute dispatches.

In short, Nadii connects demand and supply forecasts, shelf-exposure requirements, the marketing calendar and costs in one operational decision. Promotions can then drive profit instead of leaving inventory behind.

With tens of thousands of SKUs across hundreds of locations, manually tuning replenishment policies simply does not scale. Decision-makers struggle to know whether policies are being executed as intended, while operational teams struggle to explain the trade-offs between availability, transport costs, minimum volumes and deadlines. The natural temptation is to simplify the logic so calculations remain manageable — at the cost of precision and money.

Nadii was designed for this scale. At its core are cost-probabilistic algorithms developed by ML specialists who hold Kaggle Grandmaster titles. By default, the system automates repeatable work: it adjusts replenishment policies, cleans data anomalies, creates order proposals and sends them to the ERP. Where EDI is unavailable, it can generate purchase orders in the supplier's required format for email or print. MOQ, pack sizes, multiple currencies, tariffs and supplier discounts are considered automatically, and items can be consolidated into full transport units.

Nadii's engine uses computing power according to business importance. A flexible server pool allows critical products and nodes to be recalculated more frequently and with greater precision, while less important items can be processed less frequently without changing the underlying cost-probabilistic objective. This creates scale without falling back on simplistic rules, while keeping every recommendation transparent: users can see why it was generated and which trade-offs were accepted.

The algorithms consider the full range of relevant drivers: seasonality, campaigns, events, competitor prices and availability, supplier lead times and service levels, warehouse and transport constraints, marketplace rules, presentation minima and more. As a result, small changes in demand, supply or cost can quickly translate into micro-decisions about allocation, reservations, dispatch schedules, relocations and purchasing.

Recalculation frequency is configurable: daily, several times per day or every few days, depending on the category, channel and operating model. Users intervene where human judgment adds real value — new products, exceptions and strategic decisions — while the rest happens automatically in the background with a complete decision trail.

What you gain:

• scale without proportional headcount growth: more SKUs and locations without the same increase in team workload,
• fewer errors and delays because repeatable decisions happen automatically and on time,
• a shorter forecast → order → delivery cycle with fewer operational bottlenecks,
• full assortment coverage, including new products with rapid initialization of starting quantities,
• consistent ERP/WMS execution through one decision stream instead of manual data copying.

Nadii structures the assortment from portfolio level down to the individual SKU. For every SKU and location, it manages a status such as active, sell-through, order-on-demand or discontinued. These statuses can be synchronized with the ERP or maintained independently. Sales and inventory planning can therefore reflect the real commercial status of each item, while reporting shows performance and inventory movements by status to support decisions about which products should remain in the assortment.

In parallel, Nadii manages the product lifecycle: ramp-up for new products, in-season availability and controlled ramp-down. For shelf-life-sensitive items, the system signals risk early, identifies the appropriate last-order date and — where justified — proposes transfers so inventory is sold before expiry instead of being written off.

If you sell kits or sets, Nadii connects demand for the finished kit with demand for each component. It forecasts both levels, monitors inventory and proposes purchases accordingly, reducing the risk that one missing component blocks sales of the entire kit. It can also support simple assembly or picking so the purchasing plan remains aligned with kit sales.

What you gain:

• Less dead stock and fewer expiries through earlier warnings and controlled phase-out.
• Better use of shelf space and working capital through location-level statuses and visible presentation-stock cost.
• More consistent launches and seasonal transitions through clear ramp-up and ramp-down policies and the right last-order date.
• Kits without avoidable bottlenecks because component demand and kit demand are planned together.
• Less manual work and more predictability because statuses, lifecycles, reporting and alerts are managed systematically.

In short, Nadii combines statuses, lifecycle, kits and shelf-exposure rules in one decision logic. The result is a clearer assortment, more stable availability and capital allocated where it can actually generate sales.

Central warehouse, region, store or online channel — Nadii provides probabilistic forecasts at the level required for each decision. Models combine sales history with information about promotions, prices, seasonality and special events. In parallel, Nadii forecasts supply through lead-time distributions and supplier service levels so decisions reflect real uncertainty on both sides of the equation.

In the short term, the system uses confirmed orders and reservations. Over longer horizons, continuously learning models take over. Nadii monitors forecast performance and automatically switches methods when the demand regime changes — for example after the launch of a campaign or at the start of a season. Decisions are therefore not based on one deterministic number, but on probability distributions and financial consequences: shortage risk versus excess cost.

What you gain:

• stable availability with lower inventory because decisions reflect both risk and cost,
• fewer surprises through earlier signals of demand peaks or supplier-side risk,
• consistent planning from distribution center to store, with coherent aggregation up and down the network,
• better budgeting and cash-flow planning because forecasts translate directly into purchasing and inventory plans,
• more confidence during promotions and seasonal transitions because models recognize uplift, run-down and substitution effects.

Sales, purchasing and logistics often work toward different KPIs. Nadii provides a shared dashboard that explains the drivers behind performance: forecast error, deliberate purchasing into a discount through Smart Buying, supplier delays, or warehouse and transport constraints. Every recommendation has a cost rationale — lost margin, excess-inventory cost, transport cost and cash impact — together with a complete audit trail showing who changed what and why.

Data and decisions can be sent automatically to BI or a Data Lake. Roles and permissions ensure that management sees the macro picture while operations receive concrete actions, comments and alerts only where intervention genuinely adds value.

What you gain:

• one source of truth instead of parallel versions of the same KPI,
• faster decisions and fewer disputes because every team works from the same facts and costs,
• better alignment between departments through shared KPIs and a shared understanding of performance drivers,
• a more predictable impact on P&L by connecting operating decisions to margin and inventory,
• compliance and control through complete change tracking and automated reporting for C-level, managers and operations.