We consider a scheduling problem where a manufacturer processes a set of jobs for a customer and delivers the completed jobs to the customer. The job sizes and processing times are given. The objective is to minimize the maximum delivery time to the customer. In the production stage, one machine with an unavailability period is used to process the jobs. The machine has a fixed capacity and the jobs are processed in batches under the condition that the total size of the jobs in a batch cannot exceed the machine capacity. The processing time of a batch is the maximum processing time of the jobs contained in the batch. In addition, each batch is non-resumable, i.e., if the processing of a batch cannot be completed before the unavailability period, the batch needs to be processed anew after the unavailability interval. In the distribution stage, the manufacturer assigns a vehicle with a fixed capacity to deliver the completed jobs. The total size of the completed jobs in one delivery cannot exceed the vehicle capacity. We first consider the case where the jobs have the same size and arbitrary processing times, for which we provide a 3/2-approximation algorithm and show that the worst-case ratio is tight. We then consider the case where the jobs have the same processing time and arbitrary sizes, for which we provide a 5/3-approximation algorithm and show that the worst-case ratio is tight.