THE MBTA tweeted recently that it will once again be conducting track work on the Red Line between Alewife and Harvard on weekends, closing the route during that time and running substitute shuttle bus service on weekends between November 10 and December 2. Such work is important to keep the system functional and resilient, and some delays are unavoidable when the Red Line is replaced with buses, but the current shuttle system is suboptimal for the T’s operations and for passengers on the buses. These shuttles should be run as efficiently as possible, even if it means trying new service patterns and breaking out of the “we’ve always done it that way” mindset. At TransitMatters, we believe there is a more efficient shuttle system which could be implemented, which would deliver faster trips to many passengers, take buses off of narrow parkways, and allow the T to deliver the same level of service with many fewer buses, saving the agency money.

There are two major issues with how the current Harvard-Alewife shuttle works. The first stems from the pesky fact that the Red Line runs in a relatively straight line, while buses have to stay on Boston’s famously labyrinthine roadways. The buses follow the Red Line as best they can, but since the Red Line doesn’t follow roadways beyond Porter Square, the route is circuitous.

This is especially the case from Davis to Alewife: buses leave Davis, run out through Teele Square and on to Alewife Brook Parkway before looping to the Alewife station, and then return, stopping to take on passengers at a curbside bus stop outside the Davis station not designed for such volume, blocking traffic in Davis Square. A trip that is a mile as the crow flies becomes more than two as the bus meanders on narrow, congested streets and lanes and with one corner—the right from Alewife Brook Parkway to Broadway—which is so tight that most buses have to drive up on the curb to make the turn. All told, the mile from Davis to Alewife, with traffic and lights, usually takes more than 10 minutes.

The second issue is that there is relatively little traffic from the commuter-focused Alewife on weekends, and running every bus there means that buses are frequently running empty, or close to it. A train pulling into Harvard may have enough passengers that it needs to be met by three or four buses, but by the time the buses leave Davis for Alewife, each may only have a couple of passengers left, yet all the buses still make the full trip. Running these empty buses on this long portion route is an inefficient use of resources which doesn’t even serve passengers particularly well.

We believe there is a relatively easy solution which could be implemented to address both of these issues. Instead of running one long route during service disruptions, the T should run two. The first route would follow the Red Line from Harvard north past Porter, but terminate at Davis rather than continuing to Alewife. Buses would be able to lay over in the Davis busway, which is far preferable to loading on Holland Street, easing traffic in Davis Square. The second route would run from Harvard to Alewife, but would omit Davis going outbound, running via Rindge Avenue. Coming back, it would return via Fresh Pond Parkway and Concord Avenue, skipping Davis and Porter because of the one-way street network. Instead of a 25-minute ride from Alewife to Harvard, the route would take 10 or 12 minutes.

The math here is relatively simple. Today, each bus serving the route makes a round trip (or “cycle time”) from Harvard to Alewife in about 48 minutes. If there is a Red Line train every eight minutes and four buses meet each train, this service requires four buses every eight minutes, or 24 buses per hour. In this two-route proposal, each route would have a 24-minute cycle time. However, resources would be allocated more efficiently. Four buses would still meet each train, but three would run to Davis, and only one to Alewife. Since the cycle time for each bus would be half as long, the number of buses required to provide the service would be half as many.

For most passengers, this would at worst make no difference, and for many would be a better experience. From Davis and Porter to Harvard, there would be little change in service (although the shorter, simpler route would likely make the service more reliable). For passengers from Alewife, most—anyone traveling to Harvard Square to connect to the Red Line—would save 10 to 12 minutes of travel time each way. People traveling between Alewife and Davis would face a longer trip in both directions, but this is a trip taken by a small minority of passengers. Alewife-to-Porter passengers would have a longer trip inbound, but would make up for it with a significantly shorter trip from Porter to Alewife. On the whole, passengers from Alewife, most of whom are traveling to Harvard, Kendall, or beyond, would see significant time savings.

For the T, running fewer buses would save the agency money. Instead of requiring 24 buses each day, this service could operate with just 12. Assuming 18 hours of full service each day, this would save 216 service hours per day, or 1,728 hours over the course of the four planned weekend closures. According to reported data, the cost to operate an MBTA bus is $185.60; at this rate, this plan would save the MBTA more than $300,000 by early December, and perhaps more, since many of these drivers are probably working overtime shifts at extra pay. As this track work project continues, successful implementation of this schedule could save millions of dollars in coming years, and could push the T to examine other bus shuttle ideas which could benefit both passenger experience and the T’s bottom line.

Keeping the MBTA in a state of good repair is important for our region and our economy. Weekend substitute buses are a reality we have to live with, but it can be an opportunity as well. Instead of using the same old system which is slow and inefficient, we should explore ways to make the system efficient, innovative, and effective for both passengers and for the MBTA.

Ari Ofsevit is a TransitMatters board member.