Route optimization during Covid-19 crisis

The COVID-19 crisis which is affecting the world today is testing all the players in the Supply Chain. Indeed, from the initial collapse of Chinese production to the recent stock-outs due to unprecedented demand, every link is under tension. All actors must face a double challenge: ensuring their own survival by adapting to these new conditions, but also that of the whole society whose livelihoods is provided, daily, by the Supply Chain itself.

An essential link, the last mile is heavily used and the optimization tools that support it will now have to ensure the transition to more agile logistics.

What impact does Covid-19 have on route optimization ?

To operate, route optimization uses a certain amount of information which constitutes the “input data” of the problem to be solved:

  • the optimization objectives, that is to say what we are trying to optimize: this can be costs, quality of service, environmental impact, etc.
  • the cost structure impacting tours (hourly cost, fuel price, vehicle wear cost, etc.),
  • the resources available to carry out the operations: the vehicles and their characteristics,
  • activity data: the points to be delivered, their characteristics and their constraints
  • organizational and company specific constraints, for example: “my vehicles leave in 2 waves, one at 5 am and another at 10 am”,
  • exogenous data such as road traffic or regulatory constraints regarding traffic depending on the areas and vehicle type.

The health crisis that we are going through has very often upset, when it has not swept away, all these well-established notions.

For many, the optimization goals have changed: for last-mile carriers who deliver at home, it is much less crucial to respect the delivery slots at the time of compulsory quarantine. Some will choose to loosen this constraint in order to boost the objective of optimizing resources and thus deal with the absence of parts of their team.

The cost structure is no longer the same either: the drop in fuel prices, the new HR derogations which apply to transport (driving time, work on Sundays), or even the costs of equipment for protection must be taken into account since they impact the choice of optimization algorithms.

Due to the quarantine: traffic has become almost non-existent. In response to the situation, the government also temporarily lifted certain traffic bans on freight vehicles over 7.5 tonnes at certain times. Traffic models and new regulatory constraints must also be able to be updated so that route optimization remains relevant.

The importance of an agile solution

The confinement and closure of a large number of points of sale have led certain players to venture into new models and to aim at new targets to maintain their activity.

Such changes cannot be made without adapting the processes and reorganizing. It also requires adapting to new constraints, specific to the new customers served. These changes, even transient, need to be equipped with an agile and reactive solution capable of taking into account new constraints specific to the new activity on the fly while maintaining the level of performance.

As we have seen, the brutal changes brought about by the crisis require having the capacity to act on the optimization of its tours. The more significant the model changes in your organization, the more the ability of the solution and its algorithms to adapt to it. But again, not all organizations are in the same boat…

Some, whose activities and customers are recurring, used to work with fixed routes, optimized for long periods based on forecasts including contracts signed for the current period and the usual variations of the activity. For these organizations, the arrival of the Covid-19 has often shattered the relevance of a planned schedule. If these companies have been able to ensure the continuity of their activity, it is necessary to use the optimization solution to calculate new and more suitable routes. The difficulty will lie in obtaining a reliable long-term forecast in a context where uncertainty has become the norm.

A solution to this problem could be to switch to a “reactive”, more short-term and more dynamic mode of operation, by planning the activity to be treated day by day. This of course implies being equipped with a route optimization solution that allows it (i.e. knowing how to manage these different planning modes). Requiring great organizational agility, this change will in fact be contemplated only in the event of a sharp reduction in volumes. However, in some cases, this will be an excellent line of thought for the post-crisis period.Other organizations are already planning very dynamically. This is the case for companies for which the activity to be processed is large and is known only a short time in advance. If they are equipped with a route optimization solution that already works dynamically and capable of managing real time, it will be less difficult for them to adapt to the strong vagaries that the new context implies. You still need to be able to adjust the solution to the new situation (new optimization objectives, cost structure, new HR and traffic regulations, etc.).

Finally, some organizations use a segmentation to manage their operations and only use their optimization solutions to determine the order of visit of the points within a geographic sector (scheduling). This is often the case with shipping companies, the use of sectoring making it easier to organize the sorting of packages in the warehouse. For companies for which the health crisis is synonymous with reduced volumes, the sectors are no longer relevant because they generate inconsistent tours. On the contrary, if the activity has been boosted by the crisis (home deliveries for example), the sectors must be reduced so that the tours are feasible.

If the last mile optimization solution allows it, one possible approach is to use it in “simulation” mode to resize your fleet on the one hand and redefine coherent sectors on the other. This use of Machine Learning on a short and recent history allows organizations to be more responsive and gain logistical agility.

Since the beginnings of Kardinal, the adaptability of last-mile experts has always been a source of inspiration for the design of our optimization algorithms. The challenge they face today is daunting and unprecedented. Obviously, it will be on these experts that the success of the Supply Chain will be based in the days and weeks to come. However, route optimization software can, and should, be an essential support in this period, which is testing the agility and resilience of our organizations.