A delivery company completing all its Monday deliveries without trouble does not mean the rest of the week is well planned.
A vehicle used on Monday may be needed on Tuesday for a more critical shipment. An order postponed today may have to share capacity with other deliveries tomorrow. A driver who does not work on certain days, or a customer who accepts deliveries only on Wednesday, can change the task distribution for the whole week.
For this reason, optimizing each day independently is not enough for some businesses.
Multi-day route planning is the approach that evaluates, over a planning horizon spanning several days, which day each delivery will be made, which vehicles and drivers it will be assigned to and how the daily routes will be built.
In this article, we look at how multi-day planning works, which businesses it suits and which decisions gain importance compared with daily planning.
What is multi-day route planning?
Multi-day route planning is the joint evaluation of orders, vehicles, drivers and operational constraints over a period of several days or a week.
In daily route planning, the basic question is usually:
Which vehicle should go to which customer today, and in what order?
In multi-day planning, another question is added:
On which day should each task be carried out so that it suits the planning period as a whole?
This difference matters, because the day of some deliveries is fixed while others can be made within a time range.
For example, a manufacturer may not want to plan a shipment that could be delivered between Monday and Wednesday by looking only at Monday’s free capacity. Other commitments on Tuesday and Wednesday can also affect the decision.
Multi-period vehicle routing problems address this kind of planning across time in the operations research literature. The Periodic Vehicle Routing Problem (PVRP) is also one of the important problem families in which customer visits and routes are planned over several periods. [1][2]
Differences between daily and multi-day planning
The main difference between the two approaches is the time scope of the decisions evaluated.
| Criterion | Daily planning | Multi-day planning |
|---|---|---|
| Planning horizon | One day | Several days |
| Day of tasks | Usually set in advance | Choosing among suitable days can also be a decision |
| Vehicle availability | The day in question is evaluated | Availability that varies by day is evaluated |
| Driver planning | Daily working conditions | Day-by-day availability and other working conditions |
| Capacity assessment | Capacity within the day | Distribution of capacity across days |
| Optimization objective | Performance of the daily plan | Objectives across the whole planning horizon |
| Plan changes | Affect the day in question | May also affect decisions on later days |
This distinction does not mean daily planning is inadequate or unnecessary.
If the day of every delivery is set in advance, calculating each day’s routes separately can be a suitable solution for some businesses.
But when delivery days are flexible and resources differ from day to day, including the choice of day in planning can make more sense.
Which decisions are made in multi-day route planning?
1. On which day should tasks be carried out?
Not every order has the same flexibility.
One customer may have to receive their goods on Tuesday. Another may accept delivery on any day between Monday and Thursday.
So the suitable delivery days must be determined for each order.
For example:
| Order | Suitable delivery days |
|---|---|
| A | Monday |
| B | Monday or Tuesday |
| C | Tuesday, Wednesday or Thursday |
| D | Wednesday |
| E | Thursday or Friday |
In this table, postponing order B to Tuesday looks possible.
But if Tuesday’s vehicle capacity is already full, that choice may not be suitable.
So an order having several suitable days does not mean those days are equivalent operationally.
2. Which vehicles can be used on which day?
It is not always realistic to assume the whole fleet is available every day.
A vehicle may be in maintenance, reserved for other operations on certain days or have different transport characteristics.
So in multi-day planning, resource availability should be evaluated day by day.
For example, three vehicles may be available on Monday while only two are ready on Tuesday.
Deliveries postponed from Monday to Tuesday may face more limited capacity.
3. How is driver availability evaluated?
Having a vehicle does not by itself guarantee it can be used on a given day.
Driver availability, working hours, leave and task assignment rules also matter.
In businesses where drivers are matched to particular vehicles or operations in particular, these conditions may need to be evaluated together.
The scope of the mathematical model used matters here. Not every route optimization system solves working time regulations or personnel scheduling problems by itself. The relevant rules need to be defined explicitly and, where needed, handled with separate planning models.
4. How should daily routes be built?
After tasks are assigned to days, a suitable vehicle and route layout must be built for each day.
Here the constraints of daily route optimization, such as vehicle capacity, delivery time windows, service times and travel times, remain valid.
What sets multi-day planning apart is that day assignment and daily routes affect one another.
For example, moving a customer to Wednesday can help build a better route together with other Wednesday deliveries in the same area.
But this outcome is not guaranteed. Customer locations, time windows and available resources must be evaluated together.
Example: Why can filling today break the weekly plan?
One of the most important features of multi-day planning is that decisions that look locally suitable can cause problems over a wider planning horizon.
Let’s look at this with a simplified example.
Suppose a business has a single vehicle and a two-day planning horizon.
The vehicle can run at most 8 hours of operation each day.
There are three delivery jobs:
| Task | Estimated operation time | Suitable days |
|---|---|---|
| A | 6 hours | Monday or Tuesday |
| B | 5 hours | Monday only |
| C | 2 hours | Tuesday only |
To simplify the example, the durations here represent each task as an independent depot-departure-and-return operation block. Travel, service and required preparation times are assumed to be included in these blocks, and the blocks are assumed to be run one after another.
Real route durations cannot always be added up this way. Combining several deliveries into a single route can change the durations.
Approach 1: Placing tasks in order, only on the first suitable day
Suppose the planner takes task A first and assigns it to the first suitable day, Monday.
Monday:
- Task A = 6 hours
- Remaining capacity = 2 hours
But task B can only be done on Monday and needs 5 hours.
So Monday’s capacity cannot hold both tasks:
6 + 5 = 11 hours
The daily limit is 8 hours.
An approach that places tasks in order like this and does not reconsider its earlier decisions cannot schedule task B.
Approach 2: Evaluating the two days together
Now let’s consider all the suitable days of the tasks together.
Monday:
- Task B = 5 hours
Tuesday:
- Task A = 6 hours
- Task C = 2 hours
- Total = 8 hours
In this case:
- Monday’s capacity is not exceeded.
- Tuesday’s capacity is not exceeded.
- All three tasks can be carried out on a suitable day.
| Day | Assigned tasks | Usage | Remaining time |
|---|---|---|---|
| Monday | B | 5 / 8 hours | 3 hours |
| Tuesday | A + C | 8 / 8 hours | 0 hours |
13 of the 16 hours of total capacity over the two days are used.
What matters in this example is not that total capacity is sufficient but that capacity can be used on the right days.
The total capacity of both days was 16 hours from the start as well. The only thing that changed was how tasks were assigned to days.
This shows the importance of day selection decisions.
Note on the example: This table is a representative capacity planning scenario. No optimization was run on real road distances and no lowest-cost route was calculated. The second plan shown is only a day distribution that is feasible under the given assumptions.
The rarely discussed problem of multi-day planning: The value of future capacity
In a delivery business, empty vehicle capacity can look at first like an unused resource.
But future capacity is not always wasted space.
For example, if there is only one vehicle on Thursday that can carry out certain types of deliveries, filling that vehicle’s time with a task whose delivery day is flexible may cause problems later.
An important decision comes up at this point:
How much of tomorrow’s room to maneuver are we giving up to improve today’s plan?
The answer to this question cannot be given with total kilometers alone.
The remaining resource capacity by day, the delivery flexibility of tasks and demand that is not yet confirmed can also be taken into account.
Can planning flexibility be thought of as a resource?
Yes. As an operational way of evaluating, it can be useful to treat delivery-day flexibility as a planning resource.
For example, compare two tasks:
- Task A: Can be carried out on Monday, Tuesday or Wednesday.
- Task B: Can only be carried out on Wednesday.
Even if both tasks can be assigned to Wednesday, Task B has no option to move to another day.
So when Wednesday’s capacity is scarce, assigning A to an earlier day can help preserve options for B.
This is not the best choice in every case. Geographic clustering, transport cost and vehicle availability may support a different outcome.
But the basic idea is this:
That a task can be postponed does not mean postponing it to the last possible day is always right.
Multi-period route planning research considers objectives such as the suitable service days of orders, balance of daily workload and total transport cost together. [3]
An original evaluation approach: The flexibility budget
For businesses to understand the quality of their multi-day planning, we suggest looking not only at kilometers and vehicle count but also at how much room the plan leaves for change.
Let’s call this the “flexibility budget” here.
This term is an operational evaluation framework proposed for this article. It should not be presented as a single standardized scientific performance metric in the literature.
The framework rests on three indicators.
1. Available capacity per day
You can look at the difference between the planned workload and the available operation time for each day.
In simplified form:
Remaining capacity = Available operation time − Planned operation time
For example, if a vehicle’s daily available time is 8 hours and its planned work is 6 hours, the remaining time is 2 hours.
But there is no guarantee that these 2 hours can be used for any new delivery. Because of location, time window, vehicle capacity and other constraints, a task may not fit into this gap.
So remaining time is a preliminary indicator, not a guarantee of feasibility.
2. Delivery-day flexibility
You can track how many suitable days a task that has not yet been carried out still has.
For example:
- If task A has three suitable days, there are more planning options.
- If task B has only one suitable day left, the decision space is narrower.
But a high number of suitable days does not mean there is enough vehicle or driver capacity on all of them.
In a more comprehensive evaluation, the real resource availability of the days should also be taken into account.
3. Capacity density on upcoming days
Evaluating daily resource use only as an average can be misleading.
For example, suppose the capacity utilization rates for the five days of a week are as follows:
| Day | Planned capacity utilization |
|---|---|
| Monday | 60% |
| Tuesday | 65% |
| Wednesday | 100% |
| Thursday | 100% |
| Friday | 65% |
In this example, average capacity utilization is 78%.
But with no spare capacity on Wednesday and Thursday, responding to extra demand that may appear on those days can be hard.
Of course, 100% capacity utilization is not always wrong. If demand is confirmed and operating conditions are stable, this distribution can be suitable.
What needs to be evaluated is how capacity density relates to demand uncertainty and the likelihood of change.
When these three indicators are considered together, you can discuss not only whether a plan looks efficient but also how well it can adapt to changing conditions.
Are multi-day planning and periodic route planning the same?
Although these concepts are related, they do not describe the same problem in every case.
Multi-day route planning
A general approach that covers task, resource and route decisions within several days.
For example, deciding whether a one-time delivery is made on Monday or Tuesday can fall under it.
Periodic route planning (PVRP)
In classic periodic vehicle routing models, among the important decisions are how many times customers are visited in a planning period and which day combinations are suitable.
For example, if a customer must be visited twice a week, visit patterns such as Monday–Thursday or Tuesday–Friday can be evaluated.
The work of Francis, Smilowitz and Tzur covers the periodic vehicle routing problem and its various extensions in detail. [1]
Dynamic multi-period route planning
In this approach, new orders and planning information can appear over time.
For example, a plan prepared on Monday can be affected by new orders that arrive on Tuesday morning.
The 2010 research by Wen and colleagues examined this kind of dynamic multi-period planning problem in the context of a distribution business. [3]
A single vehicle trip lasting several days
This is a separate problem type.
A vehicle starting a long-distance trip on Monday and finishing it on Tuesday or Wednesday is not the same as planning routes that start and end at the depot every day together for several days.
Single trips that continue over several days may need additional conditions such as overnight stays, daily working limits and carrying the vehicle’s position over between periods.
So when choosing software, it should be made clear which operation model the phrase “multi-day planning” covers. [4]
How long should the planning horizon be?
Another important decision in multi-day planning is how many days of the future are evaluated together.
Should you use a three-day planning horizon, or build a two-week plan?
There is no single answer that applies to all businesses.
A short planning horizon can reduce the computational scope. But it may not take mandatory deliveries and capacity limits on later days into account early enough.
A longer planning horizon can offer the chance to evaluate future decisions. On the other hand, forecast orders, resource availability and other information can be more uncertain the further ahead you go.
As the problem grows, the computational load of the solution can also increase.
So when setting the planning horizon, the following should be considered together:
- How far in advance orders are known,
- The flexibility of delivery days,
- How often resource availability changes,
- Delivery commitments made to customers,
- The computing time that can be set aside for planning
The 2022 survey by Mor and Speranza covers different problem types in which routing decisions are handled together with the time dimension, and the research on them. [5]
How should the plan be re-evaluated when conditions change?
Having prepared a weekly plan does not mean operating conditions will stay fixed through the week.
New orders may arrive, vehicles may become unavailable or some deliveries may be canceled.
One approach that can be used in such cases is the rolling horizon.
In this approach, the system plans a certain future period. As new information appears, decisions that have not yet been carried out can be re-evaluated and the planning horizon is moved forward.
But there is an important operational limit here.
If a delivery day has been promised to a customer or a route has been communicated to a driver, continually changing all decisions may not be practical.
So businesses should determine which decisions can be changed and which must be kept.
For example:
- Confirmed decisions: Tasks that have been communicated to the driver and put into action.
- Changeable decisions: Future routes and assignments that have not been published yet.
This distinction should be defined according to the real workflow of the system in use.
Research on dynamic multi-period route planning examines evaluating new orders as they arrive and revisiting future routes with a rolling horizon. [3]
How is the success of a good multi-day plan measured?
No single performance indicator can explain the whole quality of planning.
Depending on the needs of the business, it can be useful to evaluate the following measures together.
| Indicator | What it tells |
|---|---|
| Total travel distance | Distance foreseen over the planning horizon |
| Total route duration | Planned operation time of vehicles |
| On-time completion of tasks | Compliance with delivery commitments |
| Resource utilization rates | Distribution of vehicle and driver capacity |
| Workload balance across days | How workload is spread over time |
| Unserved tasks | Work left outside the planning scope |
| Need for plan changes | How often the prepared plan is revised |
| Remaining daily capacity | Potential room to maneuver for future demand |
There is a distinction to watch for in particular here:
A balanced workload does not always mean the lowest total cost.
In some cases, clustering orders on certain days can allow vehicles to build better routes.
On the other hand, distributing the workload more evenly can help manage operational risk.
So objectives should be set according to the business’s priorities.
Which industries can benefit from multi-day route planning?
Manufacturing and industry
Distributing products leaving production across different delivery days may require evaluating vehicle availability and shipping calendars together.
Wholesale and distributors
Dealer deliveries may be made on certain days or in weekly visit patterns. So which customers are visited on which days is an important planning decision.
Logistics and transportation
Distributing transport jobs from different customers across days may require evaluating vehicle availability and operational constraints together.
Retail and store chains
Stores’ goods receiving days and hours can limit on which day shipments can be made.
Food and beverage
Day selection can gain importance because of frequent delivery requirements, specific acceptance hours and handling conditions for products. However, requirements such as shelf life or temperature conditions need to be modeled correctly and separately.
Fast-moving consumer goods (FMCG)
Distributing high-volume, recurring orders across different days may require evaluating vehicle capacity and the distribution calendar together.
The same model should not be assumed for all of these industries. Businesses differ in delivery frequency, resource structure and business rules.
How does Trackinia handle multi-day planning?
Trackinia is a route optimization and logistics operations management platform that handles vehicle and driver planning together with daily and multi-day operation scenarios.
The tasks, vehicles, drivers and availability conditions to be included in planning can be defined.
The aim is to evaluate conditions such as vehicle capacities, resource availability, working hours, delivery time windows and planning days together.
This approach makes not only which vehicle carries out a task but also how tasks are distributed to suitable days part of the planning process.
Plans can be reviewed on a map and a timeline.
This way, operations teams can evaluate daily routes within a wider planning horizon.
To find out whether multi-day planning suits your business, you can request a Trackinia demo based on your current delivery structure, number of resources and business rules.
Conclusion: A good plan should think not only about today but also about the days after
The core value of multi-day route planning is that it can bring the decisions of several days into the same evaluation process.
When delivery days, vehicle capacities, driver availability and other business rules are considered together, you can examine the effects that decisions that look suitable on a daily basis have over the wider planning period.
However, using a longer planning horizon alone does not guarantee a better result.
Accurate data, clearly defined objectives, feasible constraints and an operational evaluation of the results are still needed.
What matters for businesses is not only filling every day but carrying out the right tasks, with the right resources, on suitable days, within a plan that can adapt to changing conditions.
Editorial note: The example tables, the “flexibility budget” concept and the suggested evaluation indicators in this article were prepared by us for explanation. They are not experimental results taken directly from academic research or measured performance data from Trackinia.
