JIT fundamental ideas are:

1. Reduce WIP to bare minimum

2. Pull system

3. Supply of Raw Materials should be available as needed.

In fact, one of the major factors that increase the efficiency of a JIT system is that the demand is stable and predictable, contrary to stochastic inventory models. Hence, you can't theoretically combine the two.

At the start, when the JIT and the kanban system is get introduced, a policy variable is applicable based on the stochastic model. However, as the system matures this policy variable is set to 0. On the other hand, if you’re trying to introduce risk into JIT system- then the consideration of stochastic inventory model is applicable. I’m not sure what you are trying to do.

according to me it depends on the type of product and its demand pattern.

JIT is suitable only for products where demand variability is less.

The concept of just-in-time (JIT) delivery is generally incorporated into models by means of earliness and tardiness costs. In JIT handling of inventories you should consider penalty costs for amount of early and tardy deliveries.  See the below paper, it may help.

 An Integrated Production-Distribution Planning in Green Supply Chain: A multi-objective evolutionary approach.  

Hi Kizito

An enterprise(s) wide pull process can be implemented for 90% of most companies products using multiple independent, but planned, inventory locations. These buffers are sized according to lead-time demand and local variability and replenished according to a planned cycle/sequence in line with demand/consumption. Very slow moving items with high variability can be managed using time buffer or poisson. Spike demands need to be anticipated and built for. See the March/April 2015APICS magazine - "The Agile Supply Chain' and attached