The article examines how LLM4OR merges large language models with operations research to turn manufacturing and supply‑chain business language, data fields, and on‑site rules into computable optimization models, outlining its potential entry points in enterprise decision‑making and the challenges of modeling.
This article presents DeepStock, a deep reinforcement learning framework with policy regularization that integrates classic inventory heuristics, achieving 7% turnover reduction and multi‑million cost savings across millions of SKU‑warehouse pairs in Alibaba's self‑operated ecosystem.
This article analyzes Li Auto's plan to close about 100 underperforming stores in 2026, builds a three‑factor scoring model, conducts a cost‑benefit analysis, and formulates a 0‑1 integer programming optimization to identify which locations to shut while highlighting broader product‑line challenges.
This article applies mathematical modeling to analyze fire spread, evacuation dynamics, rescue resource allocation, and decision‑making in the 31‑storey Hong Kong fire, offering concrete formulas, time estimates, and practical safety recommendations for high‑rise buildings.
Amid record numbers of graduates and fierce competition in China's job market, this article explores how multi‑dimensional matching models, interaction effects, industry‑specific weightings, Bayesian salary estimation, dynamic decision thresholds, and multi‑attribute utility frameworks can help job seekers make informed, personalized employment decisions.
Using a cellular automata model on a 60×60 grid representing a typical lake-centered scenic area, the study simulates holiday visitor flows, analyzes crowd formation mechanisms, and demonstrates that dynamic reservation, multi‑entrance distribution, and real‑time density guidance can reduce peak congestion by up to 30%.
The article introduces the checkmark function—a mathematical model where a linear term rises while an inverse term falls, creating a unique minimum—and demonstrates its powerful applications across inventory management, fence design, production batch sizing, and equipment maintenance, highlighting common optimization features.
This article reviews five classic mathematical models—production scheduling, inventory management, quality control, reliability analysis, and energy optimization—detailing their formulations, common algorithms, and how they enhance efficiency, reduce costs, ensure product quality, and support sustainable industrial operations.
Model innovation thrives on six key strategies—problem transformation, variable reshaping, merging mechanisms with data, expanding objectives, multi‑agent modeling, and solvability design—each linking mathematical tools to real‑world needs to create more realistic, efficient, and robust decision‑support models.
This article explains how a luxury‑goods pickup service transformed manual order assignment for four engineers and 24 orders into an automated, cost‑effective dispatch system using VRPTW modeling, algorithm selection, and a genetic algorithm that balances travel time and order distribution.
This article explores how cutting‑edge optimization and simulation techniques empower supply‑chain planning—covering network design, inventory layout, and large‑scale scenario modeling—to reduce costs, improve efficiency, and enhance user experience through fast, scalable algorithms and AI‑driven insights.
The article presents how operations‑optimization and simulation algorithms empower supply‑chain planning—covering network design, inventory layout, and large‑scale simulation—to achieve cost reduction, efficiency gains, and enhanced user experience through advanced algorithmic solutions.
This article presents the ORLM intelligent decision‑making large model, detailing how real‑world decision problems are formalized and solved, the training difficulties and data synthesis methods, the transition from academic research to commercial platforms, and future technical improvement plans.
JD Retail’s supply‑chain team won the Daniel H. Wagner Prize by developing data‑driven inventory selection and allocation algorithms that optimize two‑tier RDC/FDC networks, improve order fulfillment rates, reduce stock‑out losses and costs, and have been deployed at scale across millions of orders.
JD Retail’s supply‑chain technology team leveraged data‑driven inventory selection and allocation algorithms—combining machine‑learning forecasting, heuristic heuristics, and an end‑to‑end optimization framework—to boost fulfillment rates, cut costs, and secure the prestigious Daniel H. Wagner Prize for operations research excellence.
The Taotian Group’s algorithm team, with Zhejiang University’s School of Management, created the E‑commerce Supply Chain Optimization Decision Support System—implemented as Taotian Inventory Optimization Engine (IOS)—which uses deep learning, robust optimization and game theory to solve NP‑hard location, inventory and transfer problems, providing warehouse and inventory plans that improve efficiency and consumer experience, earning a nomination for China Operations Research Society’s Operations Application Award.
JD's retail supply chain team presents a data‑driven framework combining inventory selection and allocation algorithms—ML‑Top‑K, Reverse‑Exclude, and an end‑to‑end multi‑task learning model—that improve local order fulfillment, reduce stockout loss and allocation costs, and have been deployed across its RDC/FDC network.
This article details JD Retail's award‑winning, data‑driven inventory selection and allocation algorithms, explains their mathematical models, heuristic and end‑to‑end learning solutions, presents experimental results on real‑world data, and quantifies the operational gains achieved after deployment.
The Zhejiang University and Ant Group collaborative project on green‑computing resource allocation and task scheduling received the Operations Application Award at ORSC 2024, highlighting its multi‑stage optimization, significant CPU and carbon savings, and related publications in top conferences.
The article presents a packaging‑material recommendation system for e‑commerce fulfillment that uses rule‑based filters and a three‑dimensional bin‑packing algorithm—both exact integer programming and fast heuristics—to select the smallest suitable box or bag, enabling cost reduction, better space utilization, and lower damage rates through applications such as box‑cutting, order merging, box‑type redesign, and AI‑driven protective‑packaging matching.
This article presents a bi‑objective inventory allocation model for guaranteed‑delivery advertising that simultaneously maximizes impressions for new orders and balances supply distribution, and introduces a fast alternating local‑search algorithm (BOLS) that outperforms popular multi‑objective evolutionary algorithms and the commercial solver Gurobi in extensive experiments.
This article explains why model validation is crucial for land‑use planning, outlines five practical validation techniques—including historical data checks, sensitivity analysis, scenario analysis, stress testing, and cross‑validation—and shows how each method helps identify risks and improve model robustness before real‑world deployment.
The two‑day workshop hosted by Zhejiang University Management School and Shanghai Jiao Tong University brings leading scholars and industry experts to present cutting‑edge research on operations research, large‑scale optimization, AI, graph neural networks, and supply‑chain applications, with detailed speaker bios, talk titles, abstracts, and registration information.
This article introduces a lightweight Go-based heuristic solver designed for a wide range of logistics optimization problems such as vehicle routing, scheduling, and warehouse planning, detailing its algorithmic repertoire, configurable architecture, and real‑world deployment cases.
This article introduces a mathematical model for designing compromise solutions in multi‑party decision making, explains the underlying multi‑objective optimization framework, presents a quadratic programming example, and discusses how adjusting indicator ranges can balance differing preferences to achieve mutually acceptable outcomes.
JD.com became the first Asian company to win the prestigious INFORMS Prize by leveraging AI‑powered, user‑centric supply chain innovations that deliver 95% of orders within 24 hours, scale assets to 153.8 billion yuan, and generate tens of thousands of daily logistics decisions.
This article explores a novel framework that merges causal inference with genetic algorithms to improve freight pricing strategies, addressing data limitations, bias, and dynamic market conditions, and demonstrates its robustness and effectiveness through extensive offline simulations and real‑world experiments.
The source presents several modeling challenges for the COMAP contest, covering ecological sex‑ratio dynamics of marine eels, submersible navigation and rescue, Great Lakes water‑level management, sustainable property insurance under climate risk, and strategies to curb illegal wildlife trade, each requiring development and analysis of quantitative models.
This article compiles a comprehensive list of evaluation and optimization model resources, covering Data Envelopment Analysis, ANP, VIKOR, grey relational analysis, linear and integer programming, and various meta‑heuristic algorithms with example implementations, providing a handy reference for scholars and practitioners.
This article compiles recent resources on evaluation and optimization models, covering fundamental concepts, data preprocessing techniques, weighting methods such as TOPSIS and entropy, as well as linear and integer programming, graph theory, network algorithms, and meta‑heuristic approaches like simulated annealing and genetic algorithms.
This article explains what an optimization solver is, defines key terminology, walks through simple equation‑based examples like the chicken‑rabbit problem, expands to real‑world scenarios such as ad allocation and employee benefits, and outlines common solution methods and how to choose appropriate solvers.
Pareto optimality, also known as Pareto efficiency, describes a state where improving any individual's outcome inevitably worsens another's, serving as a key criterion in multi‑objective optimization and decision science for evaluating trade‑offs such as maximizing profit while minimizing environmental impact.
This article revisits the century‑old Lanchester equation, highlights its shortcomings in high‑tech, asymmetric conflicts, proposes enhancements that incorporate weapon quality and intelligence, and demonstrates the updated model through a year‑long combat simulation, emphasizing both analytical insight and the human cost of war.
Using queueing theory and traffic flow equations, this article builds a mathematical model to explain why travel times during the Mid‑Autumn and National Day holidays can be three times longer than normal, analyzing highway speed, vehicle density, entry/exit delays, and sensitivity of key parameters.
This article presents a comprehensive overview of decision optimization algorithms applied to port terminal equipment coordination, detailing a real-world case study, the iECS architecture, implementation challenges across data, computation, and operations, and discusses future trends and best practices for industry deployment.
This presentation examines how intentionally delaying online decisions—illustrated through e‑commerce, ride‑hailing, and online gaming scenarios—can improve market thickness, reduce logistics costs, and lower regret, supported by a theoretical model, exponential regret bounds, and empirical evidence from JD.com data.
The article surveys how operations research and combinatorial optimization model 3‑D interior layout generation as a complex decision problem, describes an iterative optimization framework, and reviews recent AI models like LEGO‑Net and CC3D that reduce collisions but still leave fully automatic high‑quality design as an open challenge.
This article introduces core queueing theory concepts such as Kendall notation, M/M/1 and M/M/c models, Little’s Law, Erlang formulas, and demonstrates their practical application through a detailed Python simulation of a bank service system.
This article explores how causal inference techniques, including dynamic Bayesian networks and time‑series models, can be used to improve demand forecasting and replenishment strategies in inventory management, offering both theoretical concepts and practical case studies for operational decision‑making.
The DataFun Summit 2023 Decision Intelligence Forum features leading experts from Tsinghua University, Shanshu Technology, Cainiao Network, and Alibaba Cloud presenting cutting‑edge research and real‑world case studies on operations research, supply‑chain optimization, logistics, energy markets, and decision‑making algorithms.
This article introduces Didi’s real‑time online marketplace, explains its fundamental principles, network effects, and social efficiency benefits, and examines key technical areas such as mechanism design, decision intelligence, operations research, reinforcement learning, and causal inference that drive its advanced matching and dispatch strategies.
The article introduces Didi’s large‑scale ride‑hailing marketplace, explaining how mechanism design, decision intelligence, operations research, reinforcement learning, and causal inference are combined to create fair, efficient, data‑driven matching, routing, and incentive systems that tackle massive two‑sided market challenges.
This article explores how industrial data and intelligent algorithms can drive production scheduling optimization, discussing strategic significance, challenges, data‑driven algorithmic approaches, key scientific problems, and future trends in smart manufacturing, with insights from academic research and industry applications.
The article presents a comprehensive overview of Alibaba Health's supply‑chain algorithms for intelligent replenishment and allocation, detailing the overall architecture, model evolution from safety‑stock to reinforcement learning, simulation validation, multi‑objective optimization, and practical Q&A on deployment.
This article presents a collection of practical techniques for transforming seemingly non‑linear optimization models—such as those with max/min operators, absolute values, fractional objectives, binary products, and special constraints—into equivalent linear programming formulations, enabling the use of mature LP solvers.
Using proportional relationships between production cost, material mass, and packaging surface area, this model determines the reasonable factory price for a specific toothpaste pack size by solving equations derived from known prices of other pack sizes and analyzing cost components.
This article examines how to redistribute a fixed number of conference seats among three academic departments after enrollment changes, comparing proportional allocation, absolute and relative fairness metrics, and applying a Q‑value method to determine the final seat distribution for 20 or 21 seats.
This article introduces queueing theory, explains its mathematical models—including input, service, and queueing rules—covers the Kendall notation and outlines the main performance indicators such as queue length, waiting time, busy period, and utilization.
This curated list compiles recent articles on evaluation and optimization models, covering Data Envelopment Analysis, ANP, VIKOR, various grey analysis methods, competition ranking techniques, case studies, linear and integer programming, and a range of optimization algorithms with Python examples.
This curated guide compiles recent articles on evaluation and optimization models, covering concepts, preprocessing techniques, weighting methods such as TOPSIS and entropy, as well as linear/integer programming, graph theory, shortest‑path, max‑flow, simulated annealing, and genetic algorithms.
This article explains the concept of comprehensive evaluation problems, outlines the five essential elements of an evaluation system, and reviews classic multi‑criteria decision‑making methods such as TOPSIS, entropy weight, and AHP, while offering practical guidance on indicator selection, data preprocessing, model suitability, and result validation.
This article introduces Lanchester's combat models, explains their assumptions and symbols for conventional, guerrilla, and mixed warfare, provides a numerical example with plotted forces, and discusses the models' strengths, limitations, and relevance for analyzing conflict dynamics.
This article explains the input‑oriented and output‑oriented Data Envelopment Analysis (DEA) models under variable returns to scale, illustrates how efficiency frontiers are formed, demonstrates calculations with decision‑making units, and provides a practical example evaluating city efficiency using these concepts.
This article explores how decision‑intelligence and operations‑research techniques are applied across the automotive supply chain, detailing the industry’s structure, optimization methods at strategic, planning and execution levels, implementation difficulties, and practical lessons drawn from real‑world projects.
This article explains how Data Envelopment Analysis (DEA) is applied to assess the technological strength of industrial sectors, detailing indicator design principles, calculation steps, result interpretation, and actionable insights for improving productivity and competitiveness across Guangzhou’s industries.
This article explains the theory, models, and dual formulations of Data Envelopment Analysis (DEA), illustrating how it assesses the relative efficiency of decision‑making units using input‑output data, linear programming, and objective weight optimization.
This article introduces Data Envelopment Analysis (DEA) as a non‑parametric technique for estimating production functions and assessing the relative efficiency of decision‑making units, discusses its advantages over parametric methods, outlines key limitations, and explains related concepts such as production function properties and returns to scale.
This article presents a full case study that uses Data Envelopment Analysis (DEA) with Python to assess the efficiency of 13 airlines based on fleet size, fuel consumption, employee count, passenger‑miles and freight‑ton‑miles, detailing data preparation, model construction, solution steps, results and practical conclusions.
Data Envelopment Analysis (DEA), introduced in 1978 by Charnes, Cooper, and Rhodes, offers a powerful linear programming framework to assess the relative efficiency of comparable decision‑making units by evaluating multiple inputs and outputs, with extensions like C2R, C2GS2, and C2WH models handling scale and technical efficiency.
This article explains how the Analytic Network Process (ANP) removes AHP's restrictive assumptions, details its supermatrix algorithm and decision steps, and demonstrates its application through a case study evaluating cost, maintenance, durability, and car categories to identify the optimal vehicle.
This article explains how decision‑intelligence and operations‑research techniques are applied across the automotive supply chain, describing the industry structure, optimization methods at strategic, planning and execution levels, implementation difficulties, real‑world case studies, and lessons learned from a decade of data‑analysis practice.
This study models the economic feasibility of transporting Antarctic icebergs by tugboat over 9,600 km to supply fresh water for the Persian Gulf, comparing the per‑cubic‑meter cost—including vessel rental, fuel consumption, melting rate, and ship speed—to the cost of seawater desalination.
This article formulates a mathematical model to determine the optimal number of firefighters to dispatch after a forest fire, balancing reduced forest damage against rescue costs by incorporating variables such as fire spread speed, burn area, crew size, extinguishing speed, and both variable and fixed expenses, and solves for the crew size that minimizes total cost.
The article explains the distinction between deterministic and stochastic decision-making, outlines common operations‑research methods such as linear and nonlinear programming, and details cash‑flow concepts and the mathematical formulas for compound interest, present and future values, annuities, and capital recovery and storage factors.
This article models the daily profit of a pig farm by balancing feed and labor costs against decreasing market prices, derives the optimal selling day that maximizes profit, and conducts sensitivity and robustness analyses to assess how estimation errors and parameter changes affect the recommended timing.
This article models a single-mechanic repair shop as a single-server queue with exponentially distributed arrivals and uniformly distributed service times, then uses Python to simulate one workday and 1,000 workdays, reporting average daily serviced customers and average customer waiting time.
This tutorial demonstrates how to model and solve a chocolate‑production profit maximization problem using Python's PuLP library, covering decision‑variable definition, objective formulation, constraints, and result extraction in a clear step‑by‑step guide for beginners.
This article derives the theoretical model of a multi‑server queueing system, detailing Poisson arrivals, exponential service times, state balance equations, and formulas for average queue length, system size, waiting and sojourn times.
This article introduces the fundamentals of queueing theory, covering basic models, the structure of input and service processes, Kendall notation, and the primary performance indicators used to evaluate and optimize queueing systems.
This article walks through the theoretical derivation of the classic M/M/1 queueing model, detailing arrival and service rates, state balance equations, performance metrics such as utilization, average number in system, average waiting time, and average residence time, with illustrative formulas and explanations.
This article outlines the HiMCM 2020 Problem B, which challenges teams to develop models for efficiently allocating limited funds to protect endangered Florida plant species over long time horizons, detailing objectives, data fields, and suggested analytical approaches.
Discover how the Analytic Hierarchy Process (AHP) transforms qualitative and quantitative factors into a hierarchical decision model, outlining its core concepts, four-step modeling procedure, pairwise comparison matrices, consistency checks, and a concise summary of its application across various fields.
This article explains the minimum cost flow problem—delivering a specified amount of flow from supply nodes to demand nodes at the lowest possible cost—covers its linear programming formulation, highlights key solution methods such as successive shortest path, cycle canceling, primal‑dual, network simplex, and outlines its primary applications in distribution network optimization.
This article explains the fixed‑cost production planning problem, builds a mathematical model with binary and integer variables, and demonstrates its solution on a garment factory example using Python's PuLP library to maximize profit.
This article introduces the facility location problem, outlines its key concepts, elements, and objective functions, and demonstrates a practical fire‑station placement case study solved with the Python Pulp library to minimize the number of stations while ensuring response times under 15 minutes.
Linear programming, a key optimization technique, is introduced with its core concepts, components, a chocolate production example, standard formulation, and graphical solution method, illustrating how to model real-world resource allocation problems and find profit‑maximizing decisions.
This article introduces integer programming, explains its standard form, outlines two primary solution techniques—branch‑and‑bound and cutting‑plane methods—and demonstrates their application through a chocolate‑production optimization case that maximizes profit under resource constraints for the company.
The article compares non‑intervention (natural) traffic, where models autonomously maximize UV, with intervention (allocation) traffic that fine‑tunes re‑ranking to meet short‑term business goals, outlines objectives of balancing immediate profit and long‑term value, and presents two technical solutions—an ML‑plus‑OR integer‑programming model and a PID‑based control loop—for real‑time traffic allocation.
The article surveys traditional elevator dispatching methods such as FCFS, SSTF, SCAN, LOOK, and SATF, then examines real‑time strategies like EDF, SCAN‑EDF, PI, and FD‑SCAN, and concludes with a discussion of modern group‑control research and detailed system requirement analysis.
This article details Ctrip’s large‑scale intelligent workforce scheduling project, covering the business background, constraint‑rich problem modeling, heuristic algorithm selection and benchmarking, performance optimizations through multithreading and distributed computing, and the design of a scheduling platform that delivers high‑quality rosters within minutes.
This article explores elevator scheduling challenges, starting with a relatable programmer’s missed movie, then delves into traditional algorithms like FCFS, SSTF, SCAN, LOOK, and SATF, followed by real‑time strategies such as EDF, SCAN‑EDF, PI, and FD‑SCAN, and finally discusses modern group‑control research and practical system specifications.
This article explains the mathematical formulation of optimization problems, distinguishes continuous and combinatorial cases, describes how optimization software interacts with user‑defined functions, and provides extensive categorized lists of free, open‑source, proprietary, and academic‑use optimization tools.
Suning applies operations‑research‑based traffic allocation to transform saturated e‑commerce traffic into retained value, using unified data, precise flow analysis, and personalized matching of users, products and contexts to maximize efficiency and revenue.
Meituan’s intelligent delivery system integrates operations‑research, machine learning, and IoT across three layers—structural optimization, market adjustment, and real‑time matching—to plan smart areas, schedule riders, route orders, and dispatch efficiently, achieving measurable travel‑distance reductions and significant time savings.
The article explains why Didi needs advanced dispatch algorithms, describes the complexities of order‑driver matching from simple one‑to‑one cases to large‑scale bipartite matching, and introduces batch matching, supply‑demand prediction, chain dispatch, and AI‑driven optimizations that together improve global efficiency and user experience.
This article explains the concept of mathematical optimization, describes how to formulate objective functions and search for optimal inputs, and provides a comprehensive list of notable free, open‑source, and proprietary optimization software along with their licensing and application domains.
This article explains how Didi’s ride‑hailing platform evolved its dispatch system—from basic nearest‑driver assignment to sophisticated batch matching, demand‑prediction, chain dispatch, and reinforcement‑learning techniques—highlighting the operational challenges, algorithmic solutions, and the massive scale impact on user experience.
This article explains the need for intelligent replenishment in Alibaba's Retail platform, introduces inventory theory and safety‑stock models, derives mathematical formulations including a linear and integer programming model, discusses practical constraints such as vendor lead time and minimum order quantities, and outlines future forecasting work.
At the INFORMS annual meeting in Phoenix, JD.com’s chief supply‑chain scientist and a team of data scientists presented a series of talks on big‑data‑driven supply‑chain optimization, demonstrated cutting‑edge logistics technologies, and engaged with global scholars and industry leaders, highlighting JD’s research strength in operations and analytics.
Alibaba’s global math competition invites participants to tackle real‑world logistics puzzles, such as the “Fastest Delivery for Couriers” network‑flow problem, while showcasing how mathematical modeling and algorithms can boost delivery efficiency by up to 30 % and highlighting the role of AI in grading.
Meituan’s AI‑driven “super brain” dispatch system combines big‑data‑fueled preparation‑time predictions, operations‑research optimization and reinforcement‑learning‑based routing to solve a dynamic vehicle‑routing problem in seconds, cutting average delivery time from 41 to 28 minutes, lowering per‑order cost over 20 % and reducing power use by 19 %.
