Managing slow-moving item: a zero-inflated truncated normal approach for modeling demand

Abstract

This paper proposes a slow-moving management method for a system using of intermittent demand per unit time and lead time demand of items in service enterprise inventory models. Our method uses zero-inflated truncated normal statistical distribution, which makes it possible to model intermittent demand per unit time using mixed statistical distribution. We conducted numerical experiments based on an algorithm used to forecast intermittent demand over fixed lead time to show that our proposed distributions improved the performance of the continuous review inventory model with shortages. We evaluated multi-criteria elements (total cost, fill-rate, shortage of quantity per cycle, and the adequacy of the statistical distribution of the lead time demand) for decision analysis using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). We confirmed that our method improved the performance of the inventory model in comparison to other commonly used approaches such as simple exponential smoothing and Croston’s method. We found an interesting association between the intermittency of demand per unit of time, the square root of this same parameter and reorder point decisions, that could be explained using classical multiple linear regression model. We confirmed that the parameter of variability of the zero-inflated truncated normal statistical distribution used to model intermittent demand was positively related to the decision of reorder points. Our study examined a decision analysis using illustrative example. Our suggested approach is original, valuable, and, in the case of slow-moving item management for service companies, allows for the verification of decision-making using multiple criteria.

Lot-Size Models with Uncertain Demand Considering Its Skewness/Kurtosis and Stochastic Programming Applied to Hospital Pharmacy with Sensor-Related COVID-19 Data

“…The current packages available in the software [ 28 ] allow us to work with distributions belonging to the GAMLSS family. The models can be selected according to goodness-of-fit criteria when fitting real-world data, as well as by generating random numbers with arbitrary distributions of interest for theoretical or empirical research [ 12 , 29 , 30 ].…”

“…The computational routines were implemented in a non-commercial software named in its version 4.0.3. For more details of this software, see (accessed on 16 July 2021), and, for packages related to inventory models, see [ 30 , 33 , 34 , 35 , 36 ]. The computer specifications used in these experimental results are reported in Table 3 .…”

Inventory pooling decisions under demand scenarios in times of COVID-19

“…We implemented our proposal in a non-commercial software named R; see http://www.r-project.org . See Rojas, Leiva, Wanke, and Marchant (2015) , Rojas, Wanke, Coluccio, Vega-Vargas, and Huerta-Canepa ( 2020 ), Wanke, Ewbank, Leiva, and Rojas (2016) and Wanke and Leiva (2015) to visualize R applications in supply models .…”

Uncertainty Quantification of Sparse Travel Demand Prediction with Spatial-Temporal Graph Neural Networks

“…In non-DL models, Jang [8] showed that travel demand follows a zero-inflated negative binomial distribution. Rojas et al [28] also noted that using a zero-inflation model is promising for modeling intermittent travel demand. However, recent demand prediction papers sidestepped this challenge by choosing a low resolution, such as 60 minutes [11,12].…”

“…We assume that the inputs follow the ZINB distribution [9,28]. A random variable that follows NB distribution has a probability mass function 𝑓 𝑁 𝐵 as:…”