The data-driven newsvendor problem: Achieving on-target service-levels using distributionally robust chance-constrained optimization

Laan, Niels van der; Teunter, Ruud H.; Romeijnders, Ward; Kilic, Onur A.

doi:10.1016/j.ijpe.2022.108509

Cited by 15 publications

(6 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ban and Rudin [3] proposed an algorithm based on the empirical risk minimization principle, and an algorithm based on kernelweights optimization to solve the "big data" newsvendor problem. Laan et al [24] studied data driven newsvendor problem with a service-level constraint. Neghab et al [25] integrated ML and optimization problem for optimizing a newsvendor's strategy.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Data-Driven Empty Container Repositioning for Large Scale Railway Network With Fuzzy Demands

Cai

Guo

et al. 2023

IEEE Trans. Fuzzy Syst.

View full text Add to dashboard Cite

The Chinese rail network has a fast-growing number of containers in service to promote intermodal transport service and to attract more freight demand to the railway for a sustainable transport system. The repositioning of empty containers induces a significant problem for the large-scale rail network, due to the spatial imbalance between freight supply and demand. Traditionally, the empty container repositioning (ECR) problem is solved with the estimated parameters of a given demand distribution, which may difficult known to decision maker. This paper develops a data-driven framework for the ECR problem based on the large datasets available. This framework employs machine learning algorithms to forecast the supply/demand of empty containers, and to draw parameters identifying the factors that can be hardly integrated into optimization models. Based on those parameters, two optimization models for different ECR modes are then proposed to explore the optimal ECR plan. An integrated weight coefficient is introduced into the objective functions, which minimizes the total kilometers that empty containers transported. The models are solved by CPLEX after constraints conversion based on triangular fuzzy supply/demand. The numerical results based on a real-world case show that the proposed solution method can yield the optimal empty container repositioning plan. The total container-kilometer may increase after data-driven parameters are introduced, as some of the empty container repositioning may no longer follow the shortest path, while the final plan becomes more practicable and executable.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

“…Therefore, the objective function of Model 1 is (2), and the fuzzy supply and demand constraint sets are (24) and (25). The same procedure can be used on ( 7) and (8) in Model 2.…”

Section: Model Solutionmentioning

confidence: 99%

Data-Driven Empty Container Repositioning for Large Scale Railway Network With Fuzzy Demands

Cai

Guo

et al. 2023

IEEE Trans. Fuzzy Syst.

View full text Add to dashboard Cite

show abstract

“…Conceptualizing problems and problem-solving Problems and problem-solving do not hold a single universal definition (Frensch and Funke, 1995) and have been systematically explored since the beginning of the 19th century, with roots in cognitive and gestalt psychology (D'zurilla and Goldfried, 1971;Dunbar and Fugelsang, 2005;Funke, 2013;Scheerer, 1963). Contemporary definitions stem from a mixture of disciplines (Frensch and Funke, 1995), including cognitive psychology (Thorndike, 1898), engineering (Woods et al, 1997), organizational learning (Edmondson, 2018), mathematics (Polya, 2004) and computer science (Newell and Simon, 1972). Regardless of context, the central meaning of problems and problem-solving do share some common denominators independent of conditions: there is a current state, there is a desired state and there is an actor who experiences disruption and lack of knowledge about potential courses of action to get from one to the other (Frensch and Funke, 1995).…”

Section: Theorymentioning

confidence: 99%

How to facilitate manufacturing industry learning from problems: a review on advanced technology problem-solving

Mohlin

2023

JWL

View full text Add to dashboard Cite

Purpose The purpose of this paper is twofold: to identify and map contemporary research on advanced technology implementations for problem-solving purposes in the manufacturing industry, and to further understand the organizational learning possibilities of advanced technology problem-solving in the manufacturing industry. Design/methodology/approach This paper outlines a scoping review of contemporary research on the subject. The findings of the review are discussed in the light of theories of contradicting learning logics. Findings This paper shows that contemporary research on the subject is characterized by technological determinism and strong solution-focus. A discussion on the manufacturing industries’ contextual reasons for this in relation to contradicting learning logics shows that a Mode-2 problem-solving approach could facilitate further learning and expand knowledge on advanced technology problem-solving in the manufacturing industry. A research agenda with six propositions is provided. Originality/value The introduction of advanced technology implies complex effects on the manufacturing industry in general, while previous research shows a clear focus on technological aspects of this transformation. This paper provides value by providing novel knowledge on the relationship between advanced technology, problem-solving and organizational learning in the manufacturing industry.

show abstract

“…For a review of selected existing models and approaches, see, e.g., [32]. The model usually relies on the assumption that demand distribution is known; however, this assumption hardly holds in practice [39]. Therefore, below, we review the related newsvendor-like research streams emphasizing this research.…”

mentioning

confidence: 99%

“…[10] deals with the discrete version of the newsvendor problem and shows the advantages of introducing time series data modeling and forecasting in the simulation of demand distribution. Van der Laan et al [39] consider the data-driven newsvendor problem under a service-level constraint and show that existing approaches to this problem suffer from overfitting, resulting in service levels below the target service level. Liu et al [25] consider a data-driven method proposed by Ban and Rudin [5].…”

mentioning

confidence: 99%

The newsvendor problem with normal, worst-case and binomial distribution of demand: Managerial implications with examples

Brzęczek,

Hrabec

2024

JIMO

View full text Add to dashboard Cite

The paper examines the newsvendor problem with demand distributions commonly used in the literature. Optimal order convergence is checked numerically. An important contribution is that the expected profits differ considerably in nominal and relative terms when the profit-loss ratio is low valued while the demand variability is at least moderate. Missed expected profit reaches even 10% of total order cost for a mark-up that equals to the worst case break even one. The optimal order quantities are compared to counterparts derived from sales data. The main managerial implication indicates that the normal distribution solution outperforms distribution-free solutions in predicting the maximal expected profit under empirical demand. Therefore, the MaxiMin solution is recommended to be used in practice to simplify the maximal expected profit calculus and for break-even mark-up evaluation. However, it should not be used to solve the optimal order quantity because the worst-case distribution asymmetry is determined by the profit-to-loss ratio and can be contradictory to the asymmetry of sales data. Moreover, the normal distribution optimum bounded with 0.8 service level commitment shows negative expected profit under mark-up from the range of 5-20% while demand variability is low or moderate proportionally to mark-up.

show abstract

The data-driven newsvendor problem: Achieving on-target service-levels using distributionally robust chance-constrained optimization

Cited by 15 publications

References 36 publications

Data-Driven Empty Container Repositioning for Large Scale Railway Network With Fuzzy Demands

Data-Driven Empty Container Repositioning for Large Scale Railway Network With Fuzzy Demands

How to facilitate manufacturing industry learning from problems: a review on advanced technology problem-solving

The newsvendor problem with normal, worst-case and binomial distribution of demand: Managerial implications with examples

Contact Info

Product

Resources

About