In a forest product supply chain, forest fibers (commercial and low-quality woods) are practically needed to produce forest products. However, procuring and transporting the required fibers in the most cost-effective way is challenging in terms of transportation cost and unit delivery cost (Anderson and Nelson 2004; Contreras et al. 2008; Alzaqebah et al. 2018). Harvested timbers, after pre-processing operations, are often carried by trucks; either directly or indirectly to wood-yard terminals before distributing them among supply consumers (Augustynczik et al. 2016; George and Binu 2018). Spatial dispersal and a reduction in the amount of wood fibers available for harvest, a high transportation costs, an environmental concern regarding ecological paradigms and an increasing price of final products persuaded researchers to search for an optimal wood supply-chain network in forest product industries (Taskhiri et al. 2015; Chen et al. 2017). Nowadays, the development of an optimal logistic network becomes one of the significant aspects of supply-chain planning, and even a slight improvement in transportation efficiency can significantly reduce the total cost of operations (Chung et al. 2008; Chen et al. 2017). In the wood industry, given that transportation phase is the major cost of a wood supply-chain network, using an appropriate logistic network planning it is possible to minimize the overall distance, cost of the network and finally the unit delivery cost (Beaudoin et al. 2007; Chauhan et al. 2009). Such a planning approach must meet the needs of all demand centers, while using the full capacity of suppliers (Carlsson and Rönnqvist 2007; Beed et al. 2020). Forest transportation planning, both off-road and on-road, is one of the most expensive activities in wood supply-chain network which can be modeled through mathematical modeling approaches. Depending on the size of the problem, complexity of the network and the type of decisions various solution algorithms, e.g., exact solutions, heuristics and evolutionary algorithms, have so far been proposed in the literature for solving complex and challenging transportation problems (Chung and Session 2000; Aruga 2005; Gracia et al. 2014; Zamora-Cristales et al. 2015; Charkhgard et al. 2018; Jamhuri et al. 2021).
Linear programming (LP) model, as the most common operations research method, is one of the most popular and widely used to allocate limited resources optimally among competing activities, quality improvements, cost reductions and productivity increases (Macmillan and Fairweather 1988; Acuna 2018). In the context of forestry, LP has been successfully applied to timber harvesting (Öhman et al. 2011), maximum allowable timber yield (Rijal et al. 2018), spatial forest planning (McDillet al. 2016), supply chain network (Daya and Nourelfath 2018) and transportation planning (Hosseini et al., 2018) across the world. Palander and Väätäinen (2005) presented an optimization model based on linear programming scheme in which they used backhaul opportunities to minimize unloading travel distance throughout the entire wood supply-chain network. Bredstrӧm and Rӧnnavist (2006) used a mixed integer programming model for a difficult combined distribution and ship scheduling problem. The problem includes several forest products and multiple pick up and deliveries. Beaudoin et al. (2007) address plan robustness assessment taking into account multiple sources of uncertainties related mainly to supply availability and machine capacity. Chauhan et al. (2009) demonstrated a good performance level of the heuristic approach for a small-scale wood supply-chain problem, and of the branch-and-price approach for large scale problems. Zamora-Cristales et al. (2015) used ant-colony (ACO) evolutionary algorithm for the biomass supply-chain network, i.e., procurement and logistics. Their study appraised lower cost than the usual to plan the optimum network design from forest sites to wood yard terminals. Acuna (2018) developed an integrated optimization model for handling decisions on wood flow scheduling, and routing when both commercial timbers and biomass residues needed to be transported from forest coups to the mill yards from the perspective of the Australian forest industry. Shabaev et al. (2020) proposed an approach to the optimal planning of wood harvesting and timber supply from the perspective of Russian forest product industries. They used decomposition algorithm combined with heuristics for solving their operational problem. The results showed that a saving of between 5 and 10% in the total supply chain cost compared to the base-case scenario. Jamhuri et al. (2021) developed an optimization approach for routing problem in order to reduce the total cost of network by reducing the truck fleet, eliminating overtime, and increasing the efficiency of the transportation network. The real-time supply chain planning and inventory management of each potential and existing node within the network of wood supply chain is governed by product delivery prices and operations management efficiencies of each unit throughout the entire network (Shahi 2016). This will result in improving the competitiveness of industries within the global market. Azizi et al. (2008) estimated demand of wood panels by the year of 2012 in Iran and reported the consumptions of particleboard, fiberboard and mid-density fiberboard will increase by 33%, 72% and 107%, respectively. A few years later, Mohammadi Limaei et al. (2011) based on 30-year time series analysis (1979–2009) estimated that wood export is expected to be 21,000 tons per to cover the demand of forest industry across the country.
Increasing wood demands at the global level, high transportation costs and limiting domestic wood suppliers for providing industries’ raw materials are some of the practical challenges faced with major forest companies in northern Iran (Rafighi et al. 2006; Norizah et al. 2014). Parallel to this, the general demand of wood is growing from year to year, while the amount of legitimate and/or authorized wood supply productions has been intensely decreased in recent years (Rafighi et al. 2006). So, according to country's dependence on imported woods from neighboring countries (e.g., Russia (81% by value), United Arab Emirates (2.7% by value), Azerbaijan (7.50% by value) and the rest 8.8% from other countries such as south Korea, Armenia, China, Malaysia, Turkey, etc.), it is necessary to specify the optimal wood supply chain network. Menhaj (2007) concluded given decreasing and/or remove the government’s tariff on imported woods, the social benefits and wood smugglers caused by illegitimated harvesting would decrease. In addition, since 2017 the government prohibited any commercial exploitations of Iranian natural forests for a period of ten years, due to the recent forest management policies called “breathing plan “aimed at improving forest health and protecting these resources as much as possible. Deploying this policy not only was successful, but also led to extensive illegal smuggling of logs as 10–15% higher than ever before (Sotoudeh foumani et al. 2021). This policy may put increasing pressures on forest industries to re-think of their current wood supply-chain network and practices leading to reduce total cost of the delivered products and increase productivity while reducing a surge in companies demand for illegitimate supply materials.
Because of the supply shortage, uncertainty in inbound logistics (raw materials) and avoid high transportation cost, some forest companies, mainly small-scale sawmills, are purchasing woods from illegitimate sources instead of importing timbers or supplying woods from remote afforestation regions to keep production lines active and reduce their unit production cost. Illegitimate activity not only has significantly affected forest resources, but this could help spur timber and timber market change. Therefore, in order to remain competitiveness in the global market, it is necessary for forest industries to find ways to manage supply shortages and inventories of raw materials at different locations of the supply chain network (such as source, wood-yard and production plans). This therefore, helps sawmill supply team to reduce operational costs and avoid interruptions in mill’s production lines while reducing the cost of deliver products and increasing its competitiveness. For this reason, we studied the current supply chain network and analyze three set of logistic scenarios by integrating inventory management and production planning with traditional supply chain management decision models aiming to achieve lower operation costs while decreasing illegitimate harvesting from these forests. In this study, we developed an integrated solutions based on the spatial analysis and mathematical optimization to identify the optimal logistics network of a wood supply-chain network. Spatial analysis tailored to preprocess the spatial database while a production supply chain management with inventory management formulated to improve the quality of solutions. We analyzed a few logistics scenarios to explore the sensitivity of the model while removing illegitimate harvesting and integrations in the inbound logistics (supply side) for a short-term operational problem of a case study in Mazandaran’s forest industry in northern Iran.