Any alternative to Level of Service (LOS) needs to avoid the very common mistake of using a measure that discourages compact development. For example, an LOS that seeks to reduce car trips will discourage compact development, whereas an LOS that seeks to reduce per capita car trips will not.
Why is specifying per capita in measuring level of service (LOS) important? Because a hypothetical development containing 100 people that is well-designed and compact (it has smaller homes while it mixes homes with offices and shops) will still produce more car trips outside of the development than a more dispersed, car-dependent, smaller development of 50 people because the latter development would generate less car trips outside of the development. However, when two more dispersed, car-dependent developments are constructed, those two developments (with a sum total of 100 people) would generate more per capita car trips than the more compact development.
Therefore, if we do not specify per capita LOS (the larger yet more compact development has lower per capita car trips), we inadvertently encourage the more dispersed, car-dependent projects and discourage better designed and compact projects, even though the more dispersed projects will ultimately produce more car trips overall.
Measuring levels of traffic congestion is counterproductive in creating a quality, sustainable, and safe city. To be healthy, cities (at least in town centers) need to be slower in speed, safe and comfortable for all forms of travel, human-scaled, and compact. Striving for free-flowing traffic at rush hour strongly undercuts each of those exceptionally important measures of city health. Healthy city objectives are undercut largely because worrying about traffic level of service compels a city to oversize its streets and intersections, promote car travel to the detriment of all other forms of travel, and oppose compact development.