Improving the decision-making process by considering supply uncertainty – a case study in the forest value chain

ABSTRACT

Planning decisions are generally subject to some level of uncertainty. In forestry, data describing the resources available have a major impact on operations performance and productivity. This paper aims to present a method to improve decision-making in the forest supply chain by taking supply uncertainty into account using the results of data quality assessments. The case study describes the operations planning process of a Canadian forest products company dealing with an uncertain volume of wood supply. Three approaches to constructing probability distributions based on data quality are tested. Each approach offers a different level of precision: (1) a frequency distribution of accuracy, (2) a normal distribution based on average accuracy, and (3) a normal distribution based on data quality classification. Using stochastic programming to plan transport and production shows that lower costs can be achieved with a general characterisation of the data accuracy. Not considering uncertainty when planning operations leads to a significant replanning transportation cost. Using classes of data quality to include uncertainty in operations planning contributes to reducing the transportation cost from $15.90/m³ down to $15.32/m³ representing 3.6%.

KEYWORDS:

• Supply uncertainty
• stochastic programming
• production planning
• forest value chain
• sample average approximation

Acknowledgements

The authors would like to thank the FORAC Research Consortium and its partners.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.

Additional information

Funding

Our gratitude goes as well to the Natural Sciences and Engineering Research Council of Canada (NSERC) who provided funding for this research. This work was supported by the FRQNT [grant number 257659].

Notes on contributors

Vanessa Simard

Vanessa Simard is a Ph.D. candidate in industrial engineering at Université Laval with the FORAC research consortium in Quebec, Canada. She is interested in the application of operations research to support industrials in improving their decisions, especially in supply chains dealing with uncertainty.

Mikael Rönnqvist

Mikael Rönnqvist is a professor in industrial engineering at Université Laval (Québec). He currently holds a Canada Research Chair (tier 1) in Operations Research in Natural Resources. He is a member of the research organisations/networks FORAC and CIRRELT. His research interests are in the areas of industrial and practical use of Operations Research, in particular in the forest industry.

Luc LeBel

Luc LeBel is a graduate from Université Laval in forest operations. He also has a master’s degree in Engineering Administration, a Master’s of Science in forestry, and a Ph.D. in forestry and forest products from Virginia Polytechnic Institute (Virginia Tech). He is the director FOR@C (Forest to Customer), a centre of engineering and management expertise for the advancement of the Canadian forest products industry through developing new technologies and business models. His main research area is concerned with wood procurement systems analysis and global forest value chain optimisation. In 2013 he was named ‘Forest engineer of the year’ by Quebec’s order of forest engineers.

Nadia Lehoux

Nadia Lehoux is a professor in industrial engineering at the Department of mechanical engineering at Université Laval in Quebec, Canada. Her expertise encompasses the development of decision tools to better support logistics, operations, and supply chains planning.