You could create a job on Azure that scales up the throughput of your collections only at the time of day that you need it and then scale down afterward.
If you are targeting DocumentDB from .NET, this Azure article has example code that shows how to change the throughput using the .NET SDK.
The specific (C# .NET) code referenced in the article looks like this:
//Fetch the resource to be updated
Offer offer = client.CreateOfferQuery()
.Where(r => r.ResourceLink == collection.SelfLink)
.AsEnumerable()
.SingleOrDefault();
// Set the throughput to 5000 request units per second
offer = new OfferV2(offer, 5000);
//Now persist these changes to the database by replacing the original resource
await client.ReplaceOfferAsync(offer);
// Set the throughput to S2
offer = new Offer(offer);
offer.OfferType = "S2";
//Now persist these changes to the database by replacing the original resource
await client.ReplaceOfferAsync(offer);
I assume that the DocumentDB SDK for other languages would have the same feature.
Additionally, from an Azure article found here you can use PowerShell to change the service level.
$ResourceGroupName = "resourceGroupName"
$ServerName = "serverName"
$DatabaseName = "databaseName"
$NewEdition = "Standard"
$NewPricingTier = "S2"
$ScaleRequest = Set-AzureRmSqlDatabase -DatabaseName $DatabaseName - ServerName $ServerName -ResourceGroupName $ResourceGroupName -Edition $NewEdition -RequestedServiceObjectiveName $NewPricingTier