Balancing Throughput, Capital, and Battery Life in Swapping Networks

Battery swapping is often discussed as a speed problem. It is not. Speed sets the ceiling. Profitability determines whether the ceiling is ever reached.

A swapping network that optimises for throughput alone will fail on capital. One that optimises for battery life alone will fail on availability. The viable operating point sits at the intersection of all three. Finding it is not accidental. It is engineered.

A Swapping Station Is an Inventory System

A swapping station is not a charger with automation on top. It is an inventory flow system. Batteries cycle through a loop: serve a vehicle, return to the station, charge, wait for the next swap.

The mathematics of this loop are unavoidable. Throughput and inventory are linked by time. If a station serves a certain number of vehicles per hour, the number of batteries required scales directly with how long each battery takes to recharge. Faster charging reduces inventory. Slower charging increases it. Throughput may stay constant. The capital required to sustain it does not.

This is where most discussions of battery swapping oversimplify. Speed is visible. Inventory economics are not.

Throughput Is a Strategic Choice, Not Just a Technical One

Our SwappBots are designed for 1-minute autonomous battery swaps. In theory, a single station can serve 60 vehicles per hour. In practice, operating continuously at that ceiling requires large battery inventories, many chargers, a large footprint, and significant capital concentrated in one location.

Often, moderate throughput across more locations delivers better coverage, better utilisation, and better capital efficiency than pushing one site to its limit. Throughput is therefore not just a capability. It is a deployment decision.

Capital discipline follows from this. Instead of embedding precision into expensive civil infrastructure, we concentrate intelligence inside the SwappBots. Software, sensing, and robotics handle alignment and execution. The surrounding station infrastructure stays simple and frugal. This is why our station capex is materially lower than large fixed swapping installations, and why the throughput-capex ratio is the right lens for evaluating swapping network economics.

Battery Life Is an Economic Variable

Batteries are long-life assets, especially under a Battery-as-a-Service model where we own the battery and the operator owns the vehicle. How those batteries are charged, at what rate, temperature, and frequency, directly determines their useful life and replacement cycles.

Charging aggressively preserves throughput with fewer batteries but accelerates degradation. Charging gently extends battery life but may reduce availability or require higher inventory. Neither extreme is right. What matters is whether the economic value of extended battery life outweighs the cost incurred elsewhere in the system.

Battery ageing is influenced by several interacting factors: charging rate, ambient temperature, thermal management, and driving patterns, among others. I²R losses grow as internal resistance rises with age, meaning the battery works harder to accept the same charge as it ages. Optimising one variable in isolation rarely produces an optimal outcome across the system.

Predictability Reduces the Cost of Uncertainty

When demand arrival is predictable, batteries do not need to be kept fully charged at all times. Energy can be prepared closer to the moment it is needed. This reduces idle inventory, avoids unnecessary charging stress, and smooths grid load, without compromising the driver's experience at the station.

Predictability does not eliminate uncertainty. It reduces the size of the buffers required to manage it. Smaller buffers mean lower capital deployed per unit of throughput delivered.

The U-Shaped Cost Curve

When throughput, inventory, charging aggressiveness, degradation, and demand behaviour are considered together, a pattern emerges. Total system cost follows a U-shaped curve.

Too few batteries force aggressive charging and drive up replacement costs. Too many batteries drive up idle capital and infrastructure spend. Between these extremes lies a narrow operating region where throughput is protected, battery life is respected, and capital efficiency is maximised.

That balance point is not stumbled into. It is designed for, using mathematical models that account for these variables before the first station goes live. The complexity lives in the design layer. On the ground, the output is a small set of clear, actionable parameters: how fast to charge a battery, at a given station, at a given time, under given conditions.

There is a shape in mathematics called the Gömböc. It has exactly one stable resting position. No matter how it is placed, it returns to balance through geometry alone, with no external intervention. The goal in designing a swapping network is the same: not constant optimisation, but a system that naturally settles into its optimal state once deployed.

Battery swapping succeeds not because it looks fast. It succeeds when physics, economics, and operations are aligned quietly and consistently over time.

Sources & Citations

[1] ScienceDirect — Design and Optimization of Electric Vehicle Battery Swapping Stations with Integrated Storage for Enhanced Efficiency, June 2025. Inventory sizing and throughput optimisation modelling. https://www.sciencedirect.com/science/article/abs/pii/S2352152X25019243

[2] ScienceDirect — Battery Valuation and Management for Battery Swapping Station, June 2023. Life-cycle revenue and marginal degradation cost tradeoffs. https://www.sciencedirect.com/science/article/abs/pii/S0360544223015141

[3] Babar R. et al. — Operational Strategies for EV Fast-Charging and Their Impact on Power Grid and Renewable Integration, SAGE Journals, 2025. I²R losses and internal resistance growth with battery aging. https://journals.sagepub.com/doi/10.1177/01445987251352551