Hasan Hüseyin Turan

Hasan Hüseyin Turan

Auto-generated profile, based on publicly available information

Hasan Hüseyin Turan

University of New South Wales

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 March 2022 in Annals of Operations Research
Verified
Journal Article
Open access
PDF file
FKFuat Kosanoglu
Fuat Kosanoglu
MAMahir Atmis
Mahir Atmis
HTHasan Hüseyin Turan
Hasan Hüseyin Turan
Maintenance planning aims to improve the reliability of assets, prevent the occurrence of asset failures, and reduce maintenance costs associated with downtime of assets and maintenance resources (such as spare parts and workforce). Thus, effective maintenance planning is instrumental in ensuring high asset availability with the minimum cost. Nevertheless, to find such optimal planning is a nontrivial task due to the (i) complex and usually nonlinear inter-relationship between different planning decisions (e.g., inventory level and workforce capacity), and (ii) stochastic nature of the system (e.g., random failures of parts installed in assets). To alleviate these challenges, we study a joint maintenance planning problem by considering several decisions simultaneously, including workforce planning, workforce training, and spare parts inventory management. We develop a hybrid solution algorithm ( $$\mathcal {DRLSA}$$ DRLSA ) that is a combination of Double Deep Q-Network based Deep Reinforcement Learning (DRL) and Simulated Annealing (SA) algorithms. In each episode of the proposed algorithm, the best solution found by DRL is delivered to SA to be used as an initial solution, and the best solution of SA is delivered to DRL to be used as the initial state. Different from the traditional SA algorithms where neighborhood structures are selected only randomly, the DRL part of $$\mathcal {DRLSA}$$ DRLSA learns to choose the best neighborhood structure to use based on experience gained from previous episodes. We compare the performance of the proposed solution algorithm with several well-known meta-heuristic algorithms, including Simulated Annealing, Genetic Algorithm (GA), and Variable Neighborhood Search (VNS). Further, we also develop a Machine Learning (ML) algorithm (i.e., K-Median) as another benchmark in which different properties of spare parts (e.g., failure rates, holding costs, and repair rates) are used as clustering features for the ML algorithm. Our study reveals that the $$\mathcal {DRLSA}$$ DRLSA finds the optimal solutions for relatively small-size instances, and it has the potential to outperform traditional meta-heuristic and ML algorithms.


1 March 2019 in International Journal of Production Economics
Verified
Journal Article
Closed access
Andrei SleptchenkoAndrei Sleptchenko
Andrei Sleptchenko
HTHasan Hüseyin Turan
Hasan Hüseyin Turan
SPShaligram Pokharel
Shaligram Pokharel
TETarek Y. ElMekkawy
Tarek Y. ElMekkawy

6 March 2020 in International Journal of Production Research
Verified
Journal Article
Closed access
HTHasan Hüseyin Turan
Hasan Hüseyin Turan
FKFuat KOSANOGLU
Fuat KOSANOGLU
MAMahir Atmis
Mahir Atmis
The sustainability of service and manufacturing operations rely heavily on the availability of equipment and assets. High availability of assets can be achieved with effective maintenance strategies. In this direction, we study a multi-skilled workforce planning problem to establish a resilient maintenance service network for high-value assets. We improve the efficiency of the maintenance network by optimising the workforce capacity in repair shops and achieving workforce heterogeneity by cross-training. As a solution strategy, we develop a two-stage iterative heuristic algorithm. At the first stage, the set of all feasible cross-training policies is effectively and systematically searched via a state-of-art multi-thread simulated annealing (MTSA) metaheuristic to find a policy(ies) that achieves the minimum cost. Further, the developed MTSA algorithm is enhanced with the multi-neighbourhood feature to escape from local optimality and implemented via parallel programming techniques. In the second stage, workforce capacity and spare parts inventory levels are optimised for the cross-training policy found at the first stage by a queuing approximation and a greedy heuristic. The MTSA obtains the lowest cost in 91 cases out of 128 compared to genetic algorithm (GA), variable neighbourhood search (VNS), an improved single-thread simulated annealing algorithm (SA) and integer programming-based clustering (IPBC) algorithms.

1 January 2022 in European Journal of Operational Research
Verified
Journal Article
Closed access
HTHasan Hüseyin Turan
Hasan Hüseyin Turan
FJFatemeh Jalalvand
Fatemeh Jalalvand
SESondoss Elsawah
Sondoss Elsawah
MRMichael J. Ryan
Michael J. Ryan

8 October 2022 in Expert Systems with Applications
Verified
Journal Article
Closed access
IAIsmail Ali
Ismail Ali
HTHasan Hüseyin Turan
Hasan Hüseyin Turan
SESondoss Elsawah
Sondoss Elsawah