Reducing stress in the supply chain by ‘peak-shaving’ orders
Despite forecasting tools becoming more advanced, companies may not fully integrate them into their decision rules. Research, however, shows the importance of anticipating future peak-demand periods.
This decade has imposed new challenges to the logistics sector. The pandemic disrupted global supply chains on a scale unprecedented in modern times. While the industry scrambles to recover, fluctuating oil and transportation prices caused by geopolitical tensions provide little stability.
The ongoing shortage of truck drivers adds to the complexity. The world road transport organization IRU says that shortages are “spiraling out of control”, and it’s set to get worse. The growing chasm between retiring drivers and new recruits is set to trip the rate of unfilled driver positions, which the IRU says will hit 60 percent by 2026.
The challenge and opportunity for manufacturers
In this context, manufacturers struggle to find trucks to transport their goods on peak-demand days. Manufacturers typically have contracted carriers that offer guaranteed shipment capacity at a fixed rate, with anything above the agreed capacity charged at market prices. Under the on-going driver shortage and fluctuating market prices that continue to trouble the logistics sector, this can add a hefty premium to the transportation cost.
In the current context, manufacturers struggle to find trucks to transport their goods on peak-demand days
At the same time, manufacturers increasingly rely on advanced forecasting tools that better predict when future peak-demand periods occur. Clearly, this presents an opportunity to better match transportation capacity with demand.
Peaks, troughs, and trade-offs
Manufacturers thus face a challenging trade-off: using premium transportation capacity adds to the transportation costs, yet restocking in quieter periods temporarily increases inventory costs. Traditional models typically recover after peak-demand days by increasing later orders. As advanced forecasting tools become better at forecasting future peak-demand periods, manufacturers could, however, consider using their base transportation capacity before the peak-demand days.
In a new research paper published in Production and Operations Management, Esade’s Joren Gijsbrechts and co-authors Christina Imdahl (Eindhoven University of Technology), Robert N. Boute (KU Leuven and Flanders Make) and Jan A. Van Mieghem (Northwestern University) prove it is optimal to do so using a ‘Lookahead Peak-Shaving policy’. This decision rule ‘peak shaves’ orders from future high-demand periods and brings them into the current period to even out transportation orders. This reduces costs and order variability, which has positive effects for the total supply chain.
For their research, the authors worked with a manufacturer in the fast-moving consumer goods industry (FGCM). When the researchers applied the model to real manufacturer data, it reduced the logistics costs of a total of 869 items by 6.7 percent compared with the manufacturer’s own policy.
This policy ‘peak shaves’ orders from future high-demand periods and brings them into the current period
The policy also helps haulage companies to plan and fill their trucks, positively contributing to a reduction of emissions, and mitigating the ongoing shortage of drivers, which is particularly pressing on peak-demand days.
Spill-over effects to other supply chain settings
The model can easily be extended to any setting with flexible supply, the researchers say. Costs that incur an additional premium once a pre-dedicated threshold is exceeded correspond to various settings, including production systems with overtime, sourcing from multiple capacitated suppliers, or transportation planning with a spot market.
The policy itself is straightforward to implement and therefore lends itself well to adoption in practice, the researchers conclude. The intuitive insights it will generate for managers can result in substantial savings and reduce overall stress in the supply chain by leveling transportation orders.
Assistant Professor, Department of Operations, Innovation and Data Sciences at Esade
View profile- Compartir en Twitter
- Compartir en Linked in
- Compartir en Facebook
- Compartir en Whatsapp Compartir en Whatsapp
- Compartir en e-Mail
Do you want to receive the Do Better newsletter?
Subscribe to receive our featured content in your inbox.