# A Progressive Hedging Approach to Solve Harvest Scheduling Problem under Climate Change

^{1}

^{2}

^{3}

^{4}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Problem Description

#### 2.2. Scenario Generation

^{3}ha

^{−1}] harvested on each management unit. To address this problem several climate change scenarios were used, which were then transformed into growth and yield scenarios using a growth and yield model that take into account climatic data as predictor variables (i.e., a process-based model). These kind of models simulate detailed physiological processes that describe system behavior, while the empirical approaches rely on correlative relationships, which do not explicitly describe system behaviors and interactions [37].

#### 2.3. The Model

**Sets:**

^{+}: minimally infeasible clusters, $\mathrm{C}\in {{\mathsf{\Lambda}}^{+}}^{+}$

**Parameters:**

^{3}in stand i in period t per ha., under scenario S;

_{i}: area of stand i (ha.);

**Variables:**

^{3});

**Objective function:**

**Constraints:**

- Stand harvest 0–1$${{\displaystyle \sum}}_{t\text{}\u03f5\text{}T}{x}_{i,s}^{t}=1\forall i\in I,\text{}\forall s\in S;\text{}$$
- Volume harvested in period t$${{\displaystyle \sum}}_{i\text{}\in \text{}I}{a}_{i,s}^{t}{A}_{i}{x}_{i,s}^{t}={V}_{s,t}\text{},\text{}\forall t\in T,\text{}\forall s\in S;$$
- Even-flow of harvest$${V}_{s,t}-\left(1+\alpha \right){V}_{s,t-1}\text{}\le 0,\text{}\forall t\text{}\u03f5\text{}\left\{2,\text{}\dots T\right\},\text{}\forall s\in S\text{}$$$$\left(1-\alpha \right){V}_{s,t-1}-\text{}{V}_{s,t}-\le 0,\text{}\forall t\text{}\u03f5\text{}\left\{2,\text{}\dots T\right\},\text{}\forall s\in S;$$
- Demand constraints$${D}_{t}\text{}\le {V}_{s,t},\text{}\forall t\in T,\text{}\forall s\in S;$$
- Adjacency constraints$${{\displaystyle \sum}}_{i\text{}\u03f5\text{}C}{x}_{i,s}^{t}\le |C|-1,\text{}\forall C\in {\Lambda}^{+},\text{}\forall t\in T,\text{}\forall s\in S;$$
- Non-anticipativity rule$${x}_{i,s}^{t}=\text{}{x}_{i,\text{}s\prime \text{}}^{t}\forall s,\text{}{s}^{\prime}\in \text{}{N}_{t},\text{}\forall t\text{}\in \text{}T,\forall i\in I,\text{}s\text{}\ne s\prime ;$$
- Binary requirements$${x}_{i,s}^{t}\in \left\{0,1\right\},\text{}\forall i\text{}\in I,\text{}\forall t\in T,\text{}\forall s\in S.\text{}$$

#### 2.4. Progressive Hedging Approach

#### 2.4.1. Methodology

Algorithm 1. Progressive Hedging |

begin |

STEP 0 |

$k\text{}:=0$ |

${w}^{0,s}\text{}(\forall s\text{}\in \text{}S);$ |

$\overline{x}=0$ |

STEP 1 |

Solve each scenario by max: |

$ma{x}_{{\overrightarrow{x}}_{s}\text{}\in \text{}{Q}_{s}}\text{}{\displaystyle {\displaystyle \sum}_{t\text{}\u03f5\text{}T}}{\displaystyle {\displaystyle \sum}_{h\text{}\u03f5\text{}H}}{r}_{i,s}^{t}{A}_{i}{x}_{i,s}^{t};$ |

STEP 2 |

Compute the average solution in each node: |

$\overline{x}={\displaystyle {\displaystyle \sum}_{{}_{s\text{}\in S}}}\mathrm{Pr}\left(\text{s}\right)\overrightarrow{{x}_{s}}$ |

STEP 3 |

If the solutions are equal according to the criterion: |

$||\overrightarrow{x}-\overline{x}||<\epsilon ;$ |

then terminate. |

STEP 4 |

Update $\rho $ |

Update the penalty factor: |

${w}_{k,s}=\rho \left(\overrightarrow{x}-\overline{x}\right)+{w}_{k-1,s}\text{\hspace{1em}}\left(\forall s\text{}\in \text{}S\right);$ |

STEP 5 |

Solve each scenario with the penalty term: |

$ma{x}_{{\overrightarrow{x}}_{s}\text{}\in \text{}{Q}_{s}}\text{}{\displaystyle {\displaystyle \sum}_{t\text{}\u03f5\text{}T}}{\displaystyle {\displaystyle \sum}_{h\text{}\u03f5\text{}H}}{r}_{i,s}^{t}{A}_{i}{x}_{i,s}^{t}+{w}_{k,s}{\overrightarrow{x}}_{s}+\frac{\rho}{2}||\overrightarrow{x}-\overline{x}{||}^{2}$ |

STEP 6 |

Update iteration number |

k = k+1 |

Return to STEP 2 |

End |

#### 2.4.2. PH Heuristic

**Fixing Variables and Compact Model**

**Penalty Factor**

**Hot Starts**

**Dynamic Gap**

#### 2.5. Instances of the Original Problem

#### 2.6. Hardware and Software

## 3. Results

#### 3.1. Configurations

#### 3.2. Performance Comparison

## 4. Discussion and Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## References

- Alonso-Ayuso, A.; Escudero, L.F.; Guignard, M.; Quinteros, M.; Weintraub, A. Forestry management under uncertainty. Ann. Oper. Res.
**2011**, 190, 17–39. [Google Scholar] [CrossRef] - Watson, R.T.; Zinyowera, M.C.; Moss, R.H. (Eds.) Climate Change 1995: Impacts on Forests. In Contribution of Working Group II to the Second Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, UK, 1995; 878p. [Google Scholar]
- Christensen, J.H.; Hewitson, B.; Busuioc, A.; Chen, A.; Gao, X.; Held, R.; Jones, R.; Kolli, R.K.; Kwon, W.K.; Laprise, R.; et al. Regional Climate Projections: Climate Change. In The Physical Science Basis Contribution of Working group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, UK, 2007; Chapter 11; pp. 847–940. [Google Scholar]
- Davis, L.S.; Johnson, K.N.; Howard, T.; Bettinger, P. Forest Management; McGraw Hill Publishing Company: New York, NY, USA, 2001. [Google Scholar]
- Pasalodos-Tato, M.; Makinen, A.; Garcia-Gonzalo, J.; Borges, J.G.; Lämas, T.; Eriksson, O. Assessing uncertainty and risk in forest planning and decision support systems: Review of classical methods and introduction of new approaches. For. Syst.
**2013**, 22, 282–303. [Google Scholar] [CrossRef] [Green Version] - Yousefpour, R.; Jacobsen, J.B.; Thorsen, B.J.; Meilby, H.; Hanewinkel, M.; Oehler, K. A review of decision-making approaches to handle uncertainty and risk in adaptive forest management under climate change. Ann. For. Sci.
**2012**, 69, 1–15. [Google Scholar] [CrossRef] - Badilla Veliz, F.; Watson, J.P.; Weintraub, A.; Wets, R.J.B.; Woodruff, D. Stochastic optimization models in forest planning: A progressive hedging solution approach. Ann. Oper. Res.
**2015**, 232, 259–274. [Google Scholar] - Hof, J.; Pickens, J. Chance-constrained and chance-maximizing mathematical programs in renewable resource management. For. Sci.
**1991**, 38, 308–325. [Google Scholar] - Palma, C.D.; Nelson, J.D. A robust optimization approach protected harvest scheduling decisions against uncertainty. Can. J. For. Res.
**2009**, 39, 342–355. [Google Scholar] [CrossRef] - Eyvindson, K.; Kangas, A. Stochastic goal programming in forest planning. Can. J. For. Res.
**2014**, 44, 1274–1280. [Google Scholar] [CrossRef] - Kangas, A.; Hartikainen, M.; Miettinen, K. Simultaneous optimization of harvest schedule and measurement strategy. Scand. J. For. Res.
**2014**, 29, 224–233. [Google Scholar] [CrossRef] - Birge, J.R.; Louveaux, F. Introduction to Stochastic Programming, 2nd ed.; Springer Series in Operations Research and Financial Engineering; Springer: New York, NY, USA, 2011; Volume 31. [Google Scholar] [CrossRef]
- Rockafellar, R.T.; Uryasev, S. Optimization of conditional value-at-risk. J. Risk
**2000**, 2, 21–42. [Google Scholar] [CrossRef] [Green Version] - Rockafeller, R.T.; Wets, R.J.B. Scenarios and policy aggregation in optimization under uncertainty. Math. Oper. Res.
**1991**, 16, 119–147. [Google Scholar] [CrossRef] - Lasch, P.; Badeck, F.; Suckow, F.; Lindner, M.; Mohr, P. Model-based analysis of management alternatives at stand and regional level in Brandenburg (Germany). For. Ecol. Manag.
**2005**, 207, 59–74. [Google Scholar] [CrossRef] - Lindner, M.; Garcia-Gonzalo, J.; Kolstrom, M.; Green, T.; Reguera, R.; Maroschek, M.; Seidl, R.; Lexer, M.J.; Netherer, S.; Schopf, A.; et al. Impacts of Climate Change on European Forests and Options for Adaptation; Report to the European Commission Directorate-General for Agriculture and Rural Development; 2008; Available online: https://ec.europa.eu/agriculture/sites/agriculture/files/external-studies/2008/euro-forests/annex.pdf (accessed on 19 October 2019).
- Nitschke, C.; Innes, J. A tree and climate assessment tool for modelling ecosystem response to climate change. Ecol. Model.
**2008**, 210, 263–277. [Google Scholar] [CrossRef] - Seidl, R.; Rammer, W.; Jager, D.; Lexer, M. Impact of bark beetle (Ips typographus L.) disturbance on timber production and carbon sequestration in different management strategies under climate change. For. Ecol. Manag.
**2008**, 256, 209–220. [Google Scholar] [CrossRef] - Garcia-Gonzalo, J.; Borges, J.G.; Palma, J.H.N.; Zubizarreta-Gerendiain, A. A decision support system for management planning of Eucalyptus plantations facing climate change. Ann. For. Sci.
**2014**, 71, 187–199. [Google Scholar] [CrossRef] - Rammer, W.; Schauflinger, C.; Vacik, H.; Palma, J.H.N.; Garcia-Gonzalo, J.; Borges, J.; Lexer, M.J. A web-based toolbox approach to support adaptive forest management under climate change. Scand. J. For. Res.
**2014**, 29, 96–107. [Google Scholar] [CrossRef] - Eriksson, L.O. Planning under uncertainty at the forest level: A systems approach. Scand. J. For. Res.
**2006**, 21, 111–117. [Google Scholar] [CrossRef] - Garcia-Gonzalo, J.; Pais, C.; Bachmatiuk, J.; Weintraub, A. Accounting for climate change in a forest planning stochastic optimization model. Can. J. For. Res.
**2016**, 46, 1111–1121. [Google Scholar] [CrossRef] - Álvarez-Miranda, E.; Garcia-Gonzalo, J.; Ulloa-Fierro, F.; Weintraub, A.; Barreiro, S. A Multicriteria Optimization Model for Forestry Management under Climate Change Uncertainty: An application in Portugal. Eur. J. Oper. Res.
**2018**, 269, 79–98. [Google Scholar] [CrossRef] - Weintraub, A.; Church, R.L.; Murray, A.T.; Guignard, M. Forest management models and combinatorial algorithms: Analysis of state of the art. Ann. Oper. Res.
**2000**, 96, 271–285. [Google Scholar] [CrossRef] - Wets, R.J.B. On the relation between stochastic and deterministic optimization. In Control Theory, Numerical Methods and Computer Systems Modelling; Bensoussan, A., Lions, J.L., Eds.; Lecture Notes in Economics and Mathematical Systems; Springer: Berlin/Heidelberg, Germany, 1975; Volume 107, pp. 350–361. [Google Scholar] [CrossRef]
- Alonso-Ayuso, A.; Escudero, L.F.; Ortuno, M.T. BFC, a branch-and-fix coordination algorithmic framework for solving some types of stochastic pure and mixed 0–1 programs. Eur. J. Oper. Res.
**2003**, 151, 503–519. [Google Scholar] [CrossRef] - Andalaft, N.; Andalaft, P.; Guignard, M.; Magendzo, A.; Wainer, A.; Weintraub, A. A problem of forest harvesting and road building solved through model strengthening and Lagrangean relaxation. Oper. Res.
**2003**, 51, 613–628. [Google Scholar] [CrossRef] - Mulvey, J.M.; Vladimirou, H. Applying the progressive hedging algorithm to stochastic generalized networks. Ann. Oper. Res.
**1991**, 31, 399–424. [Google Scholar] [CrossRef] - Crainic, T.G.; Fu, X.; Gendreau, M.; Rei, W.; Wallace, S.W. Progressive hedging-based metaheuristics for stochastic network design. Networks
**2011**, 58, 114–124. [Google Scholar] [CrossRef] - Fan, Y.; Liu, C. Solving stochastic transportation network protection problems using the progressive hedging-based method. Netw. Spat. Econ.
**2010**, 10, 193–208. [Google Scholar] [CrossRef] - Hvattum, L.M.; Løkketangen, A. Using scenario trees and progressive hedging for stochastic inventory routing problems. J. Heuristics
**2009**, 15, 527–557. [Google Scholar] [CrossRef] - Watson, J.P.; Woodruff, D.L. Progressive hedging innovations for a class of stochastic mixed-integer resource allocation problems. Comput. Manag. Sci.
**2011**, 8, 355–370. [Google Scholar] [CrossRef] - Marques, A.F.; Borges, J.G.; Garcia-Gonzalo, J.; Lucas, B.; Melo, I. A participatory approach to design a toolbox to support forest management planning at regional level. For. Syst.
**2013**, 22, 340–358. [Google Scholar] [CrossRef] [Green Version] - Goycoolea, M.; Murray, A.; Vielma, J.P.; Weintraub, A. Evaluating approaches for solving the area restriction model in harvest scheduling. For. Sci.
**2009**, 55, 149–165. [Google Scholar] - Fontes, L.; Landsberg, J.; Tomé, J.; Tomé, M.; Pacheco, C.A.; Soares, P.; Araujo, C. Calibration and testing of a generalized process-based model for use in Portuguese eucalyptus plantations. Can. J. For. Res.
**2006**, 36, 3209–3221. [Google Scholar] [CrossRef] - Tomé, M.; Faias, S.P.; Tomé, J.; Cortiçada, A.; Soares, P.; Araújo, C. Hybridizing a stand level process-based model with growth and yield models for Eucalyptus globulus plantations in Portugal. In Proceedings of the IUFRO Conference—Eucalyptus in a Changing World, Aveiro, Portugal, 11–15 October 2004. [Google Scholar]
- Korzukhin, M.D.; Ter-Mikaelian, M.T.; Wagner, R.G. Process versus empirical models: Which approach for forest ecosystem management? Can. J. For. Res.
**1996**, 26, 879–887. [Google Scholar] [CrossRef] - Lexer, M.J.; Lexer, W.; Hasenauer, H. The use of forest models for biodiversity assessments at the stand level. For. Syst.
**2000**, 9, 297–316. [Google Scholar] - Goulding, C.J. Development of growth models for Pinus radiata in New Zealand: Experience with management and process models. For. Ecol. Manag.
**1994**, 69, 331–343. [Google Scholar] [CrossRef] - Pukkala, T.; Kellomäki, S. Anticipatory vs. adaptive optimization of stand management when tree growth and timber prices are stochastic. Forestry
**2012**, 85, 463–472. [Google Scholar] [CrossRef] [Green Version] - Albert, M.; Schmidt, M. Climate-sensitive modelling of site-productivity relationships for Norway spruce (Picea abies (L.) Karst.) and common beech (Fagus sylvatica L.). For. Ecol. Manag.
**2010**, 259, 739–749. [Google Scholar] [CrossRef] - Trasobares, A.; Zingg, A.; Walthert, L.; Bigler, C. A climate-sensitive empirical growth and yield model for forest management planning of evenaged beech stands. Eur. J. For. Res.
**2016**, 135, 263–282. [Google Scholar] [CrossRef] - Soares, P.M.; Cardoso, R.M.; Miranda, P.; Viterbo, P.; Belo-Pereira, M. Assessment of the ENSEMBLES regional climate models in the representation of precipitation variability and extremes over Portugal. J. Geophys. Res. Atmos.
**2012**, 117. [Google Scholar] [CrossRef] [Green Version] - Quinteros, M.; Alonso, A.; Escudero, L.; Guignard, M.; Weintraub, A. Una Aplicación de Programación Estocástica a un Problema de Gestión Forestal. Rev. Ing. Sist.
**2006**, 20, 67–95. [Google Scholar] - Murray, A.T. Spatial Restrictions in Harvest Scheduling. For. Sci.
**1999**, 45, 45–52. [Google Scholar] - McDill, M.E.; Rebain, S.; Braze, J. Har-vest scheduling with area-based adjacency constraints. For. Sci.
**2002**, 48, 631–642. [Google Scholar] - Goycoolea, M.; Murray, A.T.; Barahona, F.; Epstein, R.; Weintraub, A. Harvest Scheduling Subject to Maximum Area Restrictions: Exploring Exact Approaches. Oper. Res.
**2005**, 53, 490–500. [Google Scholar] [CrossRef] [Green Version] - Watson, J.P.; Woodruff, D.L.; Strip, D.R. Progressive hedging innovations for a stochastic spare parts support enterprise problem. Nav. Res. Logist.
**2007**. [Google Scholar] - Williams, B.K. Adaptive management of natural resources—Framework and issues. J. Environ. Manag.
**2011**, 92, 1346–1353. [Google Scholar] [CrossRef] [PubMed] - Eyvindson, K.; Kangas, A. Evaluating the required scenario set size for stochastic programming in forest management planning: Incorporating inventory and growth model uncertainty. Can. J. For. Res.
**2016**, 46, 340–347. [Google Scholar] [CrossRef] [Green Version]

**Figure 1.**A scenario tree consisting of four scenarios and four stages (time periods) and 10 nodes (

**1**

**a**) is shown. The scenarios are also shown in a disaggregated form (

**1**

**b**); the non-anticipativity constraints (information sets Nt) are reflected by the balloons grouping the nodes. For example, in period two (t = 2) node two shares scenarios one and two and therefore the decisions for this node (harvest or not harvest the stand) for scenarios one and two must be the same.

**Table 1.**Comparison of performance of PH algorithm (PH) versus the extended formulation with CPLEX (EF) for all the instances including adjacency restriction and demand levels D. PH gap is computed using the best bound given by CPLEX when solving EF. The results shown in this table are the averages found for each of the 15 forests generated for each instance. NO SOL means that no solution was found in more than 50% of the 15 forest repetitions.

EF | PH | Statistics | |||
---|---|---|---|---|---|

Instance | Gap [%] | Time [s] | Gap [%] | Time [s] | Saved Time [%] |

100SD0% | NO SOL | NO SOL | 1.2 | 354.5 | - |

100SD25% | 23.2 | 36,000 | 7.5 | 2458.1 | 91.7 |

100SD50% | 5.2 | 36,000 | 4.8 | 920.2 | 98.3 |

100SD75% | 4.3 | 36,000 | 3.9 | 469.3 | 98.6 |

100SD90% | 3.6 | 36,000 | 3.5 | 765.2 | 98.7 |

100SD95% | NO SOL | NO SOL | 0.9 | 226.00 | - |

100SD100% | NO SOL | NO SOL | 1.9 | 5974 | - |

200SD0% | NO SOL | NO SOL | 3.8 | 608.4 * | - |

200SD25% | 49.4 | 36,000 | 5.4 | 3963.5 | 88.9 |

200SD50% | 27.9 | 36,000 | 6.2 | 827.1 | 97.7 |

200SD75% | 6.4 | 36,000 | 6.1 | 261.7 | 98.5 |

200SD90% | 2.9 | 36,000 | 2.7 | 224.6 | 98.7 |

200SD95% | NO SOL | NO SOL | 1.6 | 5410.6 | - |

200SD100% | NO SOL | NO SOL | 1.3 | 5853.8 | - |

400SD0% | NO SOL | NO SOL | 3.4 | 3006.7 | - |

400SD25% | 6.1 | 36,000 | 5.8 | 3996.8 | 69.6 |

400SD50% | 5.2 | 36,000 | 4.9 | 1082.2 | 95 |

400SD75% | 7.2 | 36,000 | 6.8 | 514.4 | 97.1 |

400SD90% | 5.0 | 36,000 | 4.7 | 429.0 | 97.5 |

400SD95% | NO SOL | NO SOL | 2.6 | 6887.6 * | - |

400SD100% | NO SOL | NO SOL | 1.7 | 6068.4 | - |

800SD0% | NO SOL | NO SOL | 2.3 | 3109.4 | - |

800SD25% | 1.0 | 11,092 | 1.1 | 3268.6 | 66.6 |

800SD50% | 1.5 | 9029 | 1.7 | 2275.1 | 76.3 |

800SD75% | 7.8 | 36,000 | 7.5 | 1406.8 | 92.2 |

800SD90% | 8.8 | 36,000 | 8.1 | 1390.9 | 93.8 |

800SD95% | NO SOL | NO SOL | 4.7 | 9519.8 * | - |

800SD100% | NO SOL | NO SOL | 2.7 | 11,603.2 * | - |

1000SD0% | NO SOL | NO SOL | 1.8 | 4844 | - |

1000SD25% | 0.9 | 6061 | 1.0 | 1574 | 91.4 |

1000SD50% | 1.0 | 8279 | 1.3 | 5283 | 91.6 |

1000SD75% | 7.9 | 36,000 | 7.5 | 2404 | 92.9 |

1000SD90% | 8.8 | 36,000 | 8.6 | 1787 | 93.8 |

1000SD95% | NO SOL | NO SOL | 13.7 | 36,000 | - |

1000SD100% | NO SOL | NO SOL | 13.1 | 36,000 | - |

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Garcia-Gonzalo, J.; Pais, C.; Bachmatiuk, J.; Barreiro, S.; Weintraub, A.
A Progressive Hedging Approach to Solve Harvest Scheduling Problem under Climate Change. *Forests* **2020**, *11*, 224.
https://doi.org/10.3390/f11020224

**AMA Style**

Garcia-Gonzalo J, Pais C, Bachmatiuk J, Barreiro S, Weintraub A.
A Progressive Hedging Approach to Solve Harvest Scheduling Problem under Climate Change. *Forests*. 2020; 11(2):224.
https://doi.org/10.3390/f11020224

**Chicago/Turabian Style**

Garcia-Gonzalo, Jordi, Cristóbal Pais, Joanna Bachmatiuk, Susana Barreiro, and Andres Weintraub.
2020. "A Progressive Hedging Approach to Solve Harvest Scheduling Problem under Climate Change" *Forests* 11, no. 2: 224.
https://doi.org/10.3390/f11020224