Business Analytics and Predicting the Future: Insights from Lecturer Mostafa Khatami

Business analytics has moved from a back-office reporting function to one of the most in-demand skill sets in the modern economy — precisely because organizations now expect data not just to explain the past, but to anticipate the future. Few researchers illustrate that shift better than Mostafa Khatami, Senior Lecturer in Business Analytics at the University of Wollongong's School of Business, whose work sits at the intersection of operations research, statistical learning, and real-world forecasting problems in healthcare and transportation. This guide uses his academic background as a lens for understanding what business analytics actually does, and why "predicting the future" is a more precise, technical discipline than the phrase might suggest.

Who Is Mostafa Khatami?

Khatami holds a PhD in Mathematical Sciences from the University of Technology Sydney (UTS) and currently teaches and researches business analytics at the University of Wollongong (UOW), a university that has also expanded internationally through a branch campus model, including a presence at GIFT City in India. His academic specialization spans operations research, econometrics, statistical learning, and algorithm design — the core toolkit behind modern predictive and prescriptive analytics.

In 2023, Khatami received the Australian Society for Operations Research (ASOR) Rising Star Award, a national recognition given annually to early-career researchers making significant contributions to operations research in Australia. He also serves as a National Committee Member of ASOR. By the time of that award, his research record included multiple peer-reviewed publications in leading operations research and management science journals — including the European Journal of Operational Research, Computers & Operations Research, the Journal of Scheduling, and the Journal of the Operational Research Society — with several ranked among the highest tiers (A* and A) in the Australian Business Deans Council (ABDC) journal rankings.

His applied research focuses on scheduling, facility location, and supply chain problems, particularly as they arise in two sectors where getting predictions wrong has real consequences: healthcare and transportation.

What "Predicting the Future" Actually Means in Business Analytics

The phrase "predicting the future" can sound almost mystical, but in business analytics it refers to a well-defined set of techniques, generally grouped into three categories:

1. Descriptive analytics — understanding what has already happened, using historical data summaries and visualization.
2. Predictive analytics — using statistical models and machine learning to estimate what is likely to happen next, based on patterns in historical and current data.
3. Prescriptive analytics — going a step further to recommend specific actions, often using optimization techniques to determine the best response to a predicted outcome, not just the outcome itself.

Khatami's research sits heavily in the space between the second and third categories — using forecasting and stochastic modeling not just to predict outcomes, but to optimize decisions in response to that uncertainty. This is a meaningfully more advanced use of analytics than simple trend forecasting: it's the difference between predicting that demand will rise, and calculating the precise operational decision that best prepares an organization for that rise given its constraints.

A Case Study in Predictive-Meets-Prescriptive Analytics: Blood Supply Chains

One of the clearest illustrations of this approach in Khatami's research is his work on blood supply chain stability, presented at an ASOR seminar in November 2024 under the title "Substitution or emergency order? Towards blood supply chain stability." The research addresses a genuinely high-stakes forecasting and decision problem: hospitals must constantly balance fluctuating blood donation supply against fluctuating transfusion demand, with data from Australia showing a persistent imbalance for highly compatible blood types like O-negative.

The research uses a stochastic optimization model — built around a technique called sample-average approximation (SAA) — to help hospitals decide, in real time, whether to substitute a compatible blood type from existing inventory or place an emergency order. This is business analytics doing exactly what the discipline promises: not just forecasting future blood demand, but converting that forecast into an optimized, actionable decision under uncertainty.

This kind of work reflects a broader pattern across Khatami's publication record, which also includes research on coupled-task scheduling (relevant to problems like patient appointment scheduling, where two related tasks must occur with a fixed delay between them) and urban parking satisfaction and curb space management, based on a survey of hundreds of respondents in Brisbane — research aimed at informing data-driven urban transportation policy.

Why Forecasting and Optimization Go Together

A recurring theme across Khatami's body of work — and across business analytics as a discipline more broadly — is that accurate prediction is only half the value. An organization that can forecast next month's demand with perfect accuracy still needs to know what to do with that information: how much inventory to hold, how to schedule staff or equipment, or when to trigger a contingency plan like an emergency order.

This is why operations research techniques (optimization, stochastic modeling, integer programming) are so often paired with statistical forecasting in real business analytics applications. The forecasting model answers "what's likely to happen?" The optimization layer answers "given that, what should we do?" Khatami's scheduling and facility-location research — applied to healthcare and transportation settings where mistakes are costly and resources are constrained — is a working example of exactly that pairing.

Core Techniques Behind Modern Business Analytics

Drawing on the toolkit reflected in research like Khatami's, a few core technique families recur across nearly every predictive business analytics application:

1. Statistical and machine learning models — used to identify patterns in historical data and generate forecasts of future demand, behavior, or outcomes.
2. Stochastic optimization — models that explicitly account for uncertainty (rather than assuming a single fixed forecast), producing decisions that perform well across a range of likely future scenarios.
3. Scheduling and facility-location algorithms — used to allocate scarce resources (staff, equipment, physical locations) efficiently against predicted demand.
4. Econometric modeling — statistical methods rooted in economic theory, often used to understand causal relationships behind the patterns a forecasting model detects.

Together, these techniques form the backbone of what business analytics programs at universities like UOW aim to teach: not just how to run a forecast, but how to turn that forecast into a defensible, optimized business decision.

Why This Matters for Students and Professionals

For students considering a career in business analytics, research like Khatami's illustrates an important point: the discipline's real value isn't in producing a single impressive-looking prediction — it's in building models robust enough to guide real decisions when the underlying data is messy, demand is volatile, and the cost of getting it wrong is high (as in a hospital blood shortage or a transportation bottleneck). That combination of technical rigor and applied, decision-focused thinking is what distinguishes strong business analytics work from simple data reporting.

It's also a useful reminder that "predicting the future" in a business context rarely means eliminating uncertainty — it means building systems, like Khatami's stochastic blood-substitution model, that make good decisions despite that uncertainty, and that keep improving as new data arrives.

Final Thoughts

Business analytics earns its growing importance not because it can perfectly foresee the future, but because it gives organizations a structured, data-driven way to prepare for it. Mostafa Khatami's research career — from award-recognized operations research work to applied forecasting and optimization problems in healthcare and transportation — offers a concrete picture of what that looks like in practice: rigorous statistical modeling, paired with optimization techniques, aimed squarely at real operational decisions rather than prediction for its own sake. For anyone studying or working in business analytics, that pairing of prediction and action is the real skill worth developing.