*3.5. Step 5: Battery DT KPI Quantification*

Lastly, but most importantly, it is necessary to identify the benefits of the battery DT to draw light on its significance for the battery industry. Referring to the problem identified in the Section 1 of the paper that without substantial effort to describe and quantify benefits, it is challenging to suggest that the DT concept itself may be the most appropriate solution to the challenges faced by each particular industry. To support a holistic battery DT implementation in the future, both qualitative and quantitative KPIs are identified and elaborated below:

	- Effect on optimization cost due to battery DT functionalities.
	- Cost to establish data acquisition from BMS to the battery model. Here, we assume the preexisting cost of sensors installed on the BMS and the cells.
	- Cost of data storage method, i.e., cloud server, memory drive, etc.
	- Computational cost of simulating the algorithms of the battery DT.
	- Time needed for the state estimation algorithms, optimization algorithms and other battery DT functionalities
	- Time to retrieve battery data from its application and assign it to the DT
	- Speed of battery DT alignment with actual battery, i.e., total time for executing the parameter-update step.
	- Accuracy of parameter identification.
	- Accuracy of parameter-update estimation parameter identification.
	- Accuracy of state estimation
	- DT functionalities that support the battery designers (battery design optimization)
	- DT functionalities that support the battery users
	- DT functionalities that support the battery EoL handler (RUL assessment)

Note: Functionalities is a qualitative KPI. We determine the services that a DT is capable of providing its users through the battery DT functionalities. The accuracy of those services across the battery lifecycle is a KPI for evaluating the benefits of using a battery DT.
