My research asks how forecasts, which are always uncertain, can be turned into operational policies that are profitable, robust and simple enough to implement. It began in evolutionary computation, moved to forecast-informed pricing and routing for online grocery delivery, and now concentrates on multi-horizon forecasting, inventory decisions and university–industry knowledge exchange.
Mohammad Abdollahi, Xinan Yang and Michael Fairbank Annals of Operations Research · 2026
Online grocers must decide which delivery slots to offer and at what price while orders are still arriving. This paper estimates the opportunity cost of each slot more accurately by routing forecast orders alongside confirmed ones, and introduces a dynamic slot-combination strategy that uses customers’ flexibility to make routes more efficient. Both ideas work within standard time-window routing systems and are evaluated in simulations using real-world data.
Preview the published first pageMohammad Abdollahi, Xinan Yang and Michael Fairbank, 2026. First-page preview, resized from the published PDF; CC BY 4.0.
Mohammad Abdollahi, Xinan Yang, Moncef Ilies Nasri and Michael Fairbank European Journal of Operational Research · 2023 · 309(2) · 704–718
When a customer books a delivery slot, the retailer does not yet know which other orders will arrive. This paper forecasts the orders still to come and includes them in a dynamic routing model, so that slot availability and delivery charges reflect the true cost of each booking. The approach is tested on real-world attended-home-delivery data.
Conceptual mapOne signal, different operational decisions
01 / SHARED INPUTDemand forecast + uncertaintyWhat may happen, and how sure are we?
02 / INVENTORYReplenishment policyStock position · lead times · supplier constraintsExplore the industry case →
03 / DELIVERYSlot and routing policyCustomer choice · route capacity · opportunity costExplore the doctoral case →
A conceptual link across my applied and research work. The inventory and delivery cases use different data, methods and evidence; this is not a diagram of one deployed system.
01 / RESEARCH THEME
Forecasting and decision support under heterogeneous demand
Reliable, cost-aware multi-horizon forecasting for heterogeneous product portfolios; global–local selection by demand type and history; connecting uncertainty to service-level, replenishment and supplier decisions.
Demand management, pricing and routing for last-mile delivery
Decisions about delivery slots, prices and routes when orders arrive over time. Forecast orders are used to estimate the opportunity cost of each booking; profit and unnecessary driving are both considered.
Published results and case studies are separate from the synthetic code examples. Technical demonstrations →
Earlier publications
4 peer-reviewed journal articles and 1 conference paper.
Publication details follow my latest academic record, with publisher links where available. The 2026 Annals of Operations Research article is
published on 1 September 2026, open access under a CC BY 4.0 licence, and can be downloaded here.
Seyed Mohammad Hossein Hosseini Amini, Mohammad Abdollahi and Maryam Amir Haeri. Applied Intelligence, 50, 2589–2609.
Genetic programming evolves computer programs, but the search is often slow and unfocused. This paper proposes a rule-centred variant that uses an imperialist competitive algorithm to organise the evolutionary search, improving the efficiency of program synthesis.
Mohammad Abdollahi and Mahdi Aliyari Shoorehdeli. Journal of AI and Data Mining, 8(3), 331–341.
Adapts the shuffled frog-leaping metaheuristic to evolve symbolic regression models, offering an alternative to classical genetic programming for learning mathematical expressions from data.
These are working papers and are not counted above. Status will be updated as they progress.
Forecasting and decision support under heterogeneous demand
A unified forecasting framework with global and demand-type-gated local models: evidence from deployment in a small packaging business
With Ramin Raeesi and Zhen Zhu. The earlier draft evaluated a historical 1,474-SKU portfolio at one-step horizon. The revision is developing six non-overlapping 30-day horizon evaluation and inventory-focused model selection. The separate 1,069-SKU retrospective operational result is not a peer-reviewed finding.
In preparation — multi-horizon revision
Demand management, pricing and routing for last-mile delivery
Enhancing Opportunity Cost Approximation through Incorporation of Delivery Price Reduction
Extends the doctoral work on attended home delivery by including the effect of delivery-price reductions in the estimate of opportunity cost.
Under revision
Research agenda
The next programme of work concerns how organisations move from increasingly powerful predictions to reliable
decisions. Four strands follow directly from what I have already done:
Scalable forecasting for heterogeneous demand: when global models trained across a whole
product range beat local models, when they do not, and how to decide per product without manual tuning.
Reliability-aware machine learning: forecasts that report their own uncertainty in a form an
ordering or pricing rule can use, and fallbacks that fail safely when data is thin.
Forecast-driven optimisation: pricing, demand management and inventory policies that take
forecast uncertainty as an input rather than a footnote.
Human-in-the-loop decision support: how buyers and planners actually use recommendations,
when they override them, and how systems should be designed so that overrides improve the model rather
than bypass it.
I am interested in collaboration with researchers in operations, analytics and information systems, and with
companies willing to host applied projects. The Knowledge Transfer Partnership model worked well for me and
I would gladly repeat it.
