In APO you can set for resources that their finite capacity is calculated based on their Bucket capacity settings.
You also have the possibility to set a Factor %, to adjust how much from a Bucket unit effectively can be used as available capacity.
For example you set Bucket Schema to Day with a start from 00:00, and Factor to 100%:
With factor 100, you have a continous capacity for the adjacent buckets, in our example the following days will be adjacent so if you have a capacity requirement bigger than one bucket, it still can use this capacity.
But what happens if we set capacity vector less than 100%?
For example if you set it to 75%, the capacity of 24 Hours of the Bucket Unit (Day) will be 18 Hours, so you will have capacity from 00:00 to 18:00 on each of the Day units.
Which means, if you have an unbreakable capacity requirement bigger than 18:00 hours it cannot be scheduled for thiese buckets.
What you can do is to use BAdI /SAPAPO/CRESBUCKET. With using the BAdI you have the possibility to change the pre-defined schema, for example you can group 2 days together. If you have usually requirements bigger than one day, but less than two days maybe this can be a solution for you.
Or you can make two days following each other adjacent, by adjusting Bucket of the first day to start (based on our example) at 6:00, so with the 75% Factor it will last until 0:00, and the next day's bucket will be adjacent. With that you reach continous 2 days capacity ( effectively 36 Hours with the 75% factor).
This also means that you cannot schedule bigger capacity requirement than the effective capacity of two Bucket Schema unit, if the factor is less than 100%.
If possible, and you dont want to use the BAdI mentioned, you could use a Bucket schema with bigger capacity units, like week, or month.