Nonparametric Learning Algorithms for Joint Pricing and Inventory Control with Lost-Sales and Censored Demand PDF Download

Author: Boxiao Chen
Publisher:
ISBN:
Category :
Languages : en
Pages : 44

View

Book Description
We consider a joint pricing and inventory control problem in which the customer's response to selling price and the demand distribution are not known a priori. Unsatisfied demand is lost and unobserved, and the only available information for decision-making is the observed sales data (a.k.a. censored demand). Conventional approaches, such as stochastic approximation, online convex optimization, and continuum-armed bandit algorithms, cannot be employed since neither the realized values of the profit function nor its derivatives are known. A major challenge of this problem lies in that the estimated profit function constructed from observed sales data is multimodal in price. We develop a nonparametric spline approximation based learning algorithm. The algorithm separates the planning horizon into a disjoint exploration phase and an exploitation phase. During the exploration phase, the price space is discretized, and each price is offered an equal number of periods together with a pre-specified target inventory level. Based on the sales data collected on these prices, a spline approximation of the demand-price function is constructed, and then the corresponding surrogate optimization problem is solved on a sparse grid to obtain a pair of recommended price and target inventory level. During the exploitation phase, the algorithm implements the recommended strategies. We establish a (nearly) square-root regret rate, which (almost) matches the theoretical lower bound.

Author: Boxiao Chen
Publisher:
ISBN:
Category :
Languages : en
Pages : 0

View

Book Description
We study the fundamental model in joint pricing and inventory replenishment control under the learning-while-doing framework, with T consecutive review periods and the firm not knowing the demand curve a priori. At the beginning of each period, the retailer makes both a price decision and an inventory order-up-to level decision, and collects revenues from consumers' realized demands while suffering costs from either holding unsold inventory items, or lost sales from unsatisfied customer demands. We make the following contributions to this fundamental problem as follows:1. We propose a novel inversion method based on empirical measures to consistently estimate the difference of the instantaneous reward functions at two prices, directly tackling the fundamental challenge brought by censored demands, without raising the order-up-to levels to unnaturally high levels to collect more demand information. Based on this technical innovation, we design bisection and trisection search methods that attain an O(T^{1/2}) regret, assuming the reward function is concave and only twice continuously differentiable.2. In the more general case of non-concave reward functions, we design an active tournament elimination method that attains O(T^{3/5}) regret, based also on the technical innovation of consistent estimates of reward differences at two prices.3. We complement the O(T^{3/5}) regret upper bound with a matching Omega(T^{3/5}) regret lower bound. The lower bound is established by a novel information-theoretical argument based on generalized squared Hellinger distance, which is significantly different from conventional arguments that are based on Kullback-Leibler divergence. This lower bound shows that no learning-while-doing algorithm could achieve O(T^{1/2}) regret without assuming the reward function is concave, even if the sales revenue as a function of demand rate or price is concave.Both the upper bound technique based on the "difference estimator" and the lower bound technique based on generalized Hellinger distance are new in the literature, and can be potentially applied to solve other inventory or censored demand type problems that involve learning.

Author: Huanan Zhang
Publisher:
ISBN:
Category :
Languages : en
Pages : 45

View

Book Description
We consider a periodic-review single-product inventory system with lost-sales and positive lead times under censored demand. In contrast to the classical inventory literature, we assume the firm does not know the demand distribution a priori, and makes adaptive inventory ordering decision in each period based only on the past sales (censored demand) data. The standard performance measure is regret, which is the cost difference between a feasible learning algorithm and the clairvoyant (full-information) benchmark. When the benchmark is chosen to be the (full-information) optimal base-stock policy, Huh et al. [Mathematics of Operations Research 34(2): 397-416 (2009)] developed a nonparametric learning algorithm with a cubic-root convergence rate on regret. An important open question is whether there exists a nonparametric learning algorithm whose regret rate matches the theoretical lower bound of any learning algorithms. In this work, we provide an affirmative answer to the above question. More precisely, we propose a new nonparametric algorithm termed the simulated cycle-update policy, and establish a square-root convergence rate on regret, which is proven to be the lower bound of any learning algorithms. Our algorithm uses a random cycle-updating rule based on an auxiliary simulated system running in parallel, and also involves two new concepts, namely, the withheld on-hand inventory and the double-phase cycle gradient estimation. The techniques developed are effective for learning a stochastic system with complex systems dynamics and lasting impact of decisions.

Author: Jing-Sheng J. Song
Publisher: Edward Elgar Publishing
ISBN: 180037710X
Category : Technology & Engineering
Languages : en
Pages : 565

View

Book Description
This comprehensive Handbook provides an overview of state-of-the-art research on quantitative models for inventory management. Despite over half a century’s progress, inventory management remains a challenge, as evidenced by the recent Covid-19 pandemic. With an expanse of world-renowned inventory scholars from major international research universities, this Handbook explores key areas including mathematical modelling, the interplay of inventory decisions and other business decisions and the unique challenges posed to multiple industries.

Author: Xi Chen
Publisher: Springer Nature
ISBN: 3031019261
Category : Business & Economics
Languages : en
Pages : 444

View

Book Description
This book examines recent developments in Operations Management, and focuses on four major application areas: dynamic pricing, assortment optimization, supply chain and inventory management, and healthcare operations. Data-driven optimization in which real-time input of data is being used to simultaneously learn the (true) underlying model of a system and optimize its performance, is becoming increasingly important in the last few years, especially with the rise of Big Data.

Author: Huanan Zhang
Publisher:
ISBN:
Category :
Languages : en
Pages : 0

View

Book Description
We develop the first nonparametric learning algorithm for periodic-review perishable inventory systems. In contrast to the classical perishable inventory literature, we assume that the firm does not know the demand distribution a priori and makes replenishment decision in each period based only on the past sales (censored demand) data. It is well-known that even with complete information about the demand distribution a priori, the optimal policy for this problem does not possess a simple structure. Motivated by the studies in the literature showing that base-stock policies perform near-optimal in these systems, we focus on finding the best base-stock policy. We first establish a convexity result, showing that the total holding, lost-sales and outdating cost is convex in the base-stock level. Then, we develop a nonparametric learning algorithm that generates a sequence of order-up-to levels whose running average cost converges to the cost of the optimal base-stock policy. We establish a square-root convergence rate of the proposed algorithm, which is the best possible. Our algorithm and analyses require a novel method for computing a valid cycle subgradient and the construction of a bridging problem, which significantly departs from previous studies.

Author: Hanzhang Qin (S. M.)
Publisher:
ISBN:
Category :
Languages : en
Pages : 96

View

Book Description
The thesis studies the classical multi-period joint pricing and inventory control problem in a data-driven setting. In the problem, a retailer makes periodic decisions of the prices and inventory levels of an item that the retailer wishes to sell. The objective is to match the inventory level with a random demand that depends on the price in each period, while maximizing the expected profit over finite horizon. In reality, the demand functions or the distribution of the random noise are usually unavailable, whereas past demand data are relatively easy to collect. A novel data-driven nonparametric algorithm is proposed, which uses the past demand data to solve the joint pricing and inventory control problem, without assuming the parameters of the demand functions and the noise distributions are known. Explicit sample complexity bounds are given, on the number of data samples needed to guarantee a near-optimal profit. A simulation study suggests that the algorithm is efficient in practice.

Author: Hao Yuan
Publisher:
ISBN:
Category :
Languages : en
Pages : 0

View

Book Description
We consider a periodic-review single-product inventory system with fixed cost under censored demand. Under full demand distributional information, it is well-known that the celebrated $(s,S)$ policy is optimal. In this paper, we assume the firm does not know the demand distribution a priori, and makes adaptive inventory ordering decision in each period based only on the past sales (a.k.a. censored demand) data. The standard performance measure is regret, which is the cost difference between a feasible learning algorithm and the clairvoyant (full-information) benchmark. Compared with prior literature, the key difficulty of this problem lies in the loss of joint convexity of the objective function, due to the presence of fixed cost. We develop a nonparametric learning algorithm termed the $( delta, S)$ policy that combines the powers of stochastic gradient descent, bandit controls, and simulation-based methods in a seamless and non-trivial fashion. We prove that the cumulative regret is $O( log T sqrt{T})$, which is provably tight up to a logarithmic factor. We also develop several technical results that are of independent interest. We believe that the framework developed could be widely applied to learning other important stochastic systems with partial convexity in the objectives.

Author: Guillermo Gallego
Publisher: Springer
ISBN: 1493996061
Category : Business & Economics
Languages : en
Pages : 336

View

Book Description
“There is no strategic investment that has a higher return than investing in good pricing, and the text by Gallego and Topaloghu provides the best technical treatment of pricing strategy and tactics available.” Preston McAfee, the J. Stanley Johnson Professor, California Institute of Technology and Chief Economist and Corp VP, Microsoft. “The book by Gallego and Topaloglu provides a fresh, up-to-date and in depth treatment of revenue management and pricing. It fills an important gap as it covers not only traditional revenue management topics also new and important topics such as revenue management under customer choice as well as pricing under competition and online learning. The book can be used for different audiences that range from advanced undergraduate students to masters and PhD students. It provides an in-depth treatment covering recent state of the art topics in an interesting and innovative way. I highly recommend it." Professor Georgia Perakis, the William F. Pounds Professor of Operations Research and Operations Management at the Sloan School of Management, Massachusetts Institute of Technology, Cambridge, Massachusetts. “This book is an important and timely addition to the pricing analytics literature by two authors who have made major contributions to the field. It covers traditional revenue management as well as assortment optimization and dynamic pricing. The comprehensive treatment of choice models in each application is particularly welcome. It is mathematically rigorous but accessible to students at the advanced undergraduate or graduate levels with a rich set of exercises at the end of each chapter. This book is highly recommended for Masters or PhD level courses on the topic and is a necessity for researchers with an interest in the field.” Robert L. Phillips, Director of Pricing Research at Amazon “At last, a serious and comprehensive treatment of modern revenue management and assortment optimization integrated with choice modeling. In this book, Gallego and Topaloglu provide the underlying model derivations together with a wide range of applications and examples; all of these facets will better equip students for handling real-world problems. For mathematically inclined researchers and practitioners, it will doubtless prove to be thought-provoking and an invaluable reference.” Richard Ratliff, Research Scientist at Sabre “This book, written by two of the leading researchers in the area, brings together in one place most of the recent research on revenue management and pricing analytics. New industries (ride sharing, cloud computing, restaurants) and new developments in the airline and hotel industries make this book very timely and relevant, and will serve as a critical reference for researchers.” Professor Kalyan Talluri, the Munjal Chair in Global Business and Operations, Imperial College, London, UK.

Author: Anupam Mazumdar (S. M.)
Publisher:
ISBN:
Category :
Languages : en
Pages : 81

View

Book Description
Price management, production planning and inventory control are important determinants of a firm's profitability. The intense competition brought about by rapid innovation, lean manufacturing time and the internet revolution has compelled firms to adopt a dynamic strategy that involves complex interplay between pricing and production decisions. In this thesis we consider some of these problems and develop computationally efficient algorithms that aim to tackle and optimally solve these problems in a finite amount of time. In the first half of the thesis we consider the joint pricing and inventory control problem in a deterministic and multiperiod setting utilizing the popular log linear demand model. We develop four algorithms that aim to solve the resulting profit maximization problem in a finite amount of time. The developed algorithms are then tested in a variety of settings ranging from small to large instances of trial data. The second half of the thesis deals with setting prices effectively when the customer demand is assumed to follow the multinomial logit demand model, which is the most popular discrete choice demand model. The profit maximization problem (even in the absence of constraints) is non-convex and hard to solve. Despite this fact we develop algorithms that compute the optimal solution efficiently. We test the algorithms we develop in a wide variety of scenarios from small to large customer segment, with and without production/inventory constraints. The last part of the thesis develops solution methods for the joint pricing and inventory control problem when costs are linear and demand follows the multinomial logit model.