Abstract:
An elevator system employing a micro-processor-based group controller (FIG. 2) communicating with the cars (3, 4) to assign cars to hall calls based on a Relative System Response (RSR) approach. However, rather than using unvarying bonuses and penalties, the assigned bonuses and penalties are varied using "artificial intellience" techniques based on combined historic and real time traffic predictions to predict the number of people behind a hall call, and, calculating and using the average boarding and de-boarding rates at "en route" stops, and the expected car load at the hall call floor. Prediction of the number of people waiting behind hall calls for a few minute intervals are made using traffic levels measured during the past few time intervals on that day as real time predictors, using a linear exponential smoothing model, and traffic levels measured during similar time intervals on previous similar days as historic traffic predictors, using a single exponential smoothing model. The remaining capacity in the car at the hall call floor is matched to the waiting queue using a hall call mismatch penalty. The car stop and hall stop penalties are varied based on the number of people behind the hall call and the variable dwell times at "en route" stops. The stopping of a heavily loaded car to pick up a few people is penalized using a car load penalty. These enhancements to RSR result in equitable distribution of car stops and car loads, thus improving handling capacity and reducing waiting and service times.
Abstract:
An elevator control system employing a micro-processor-based group controller (FIG. 2), which communicates with the cars (3, 4) of the system to determine the conditions of the cars, and responds to hall calls registered at a plurality of landings in the building serviced by the cars under control of the group controller, assigning hall calls to cars based on the summation for each car, relative to each call, a weighted summation of a plurality of system response factors, some indicative, and some not, of conditions of the car irrespective of the call being assigned, assigning varying "bonuses" and "penalties" to them in the weighted summation. "Artificial intelligence" techniques are used to predict traffic levels and any crowd build up at various floors to better assign one or more cars to the "crowd" predicted floors, either parking them there, if they were empty, or more appropriately assigning car(s) to the hall calls. Traffic levels at various floors are predicted by collecting passengers and car stop counts in real time and using real time and historic prediction for the traffic levels, with single exponential smoothing and/or linear exponential smoothing. Predicted passenger arrival counts are used to predict any crowd at fifteen second intervals at floors where significant traffic is predicted. Crowd prediction is then adjusted for any hall call stops made and the number of passengers picked up by the cars. The crowd dynamics are matched to car assignment, with one or more cars being sent to crowded floor(s).
Abstract:
An elevator system, and method of operating same, having a plurality of elevator cars for serving hall calls registered from the floors of a building. All of the up and down service directions from the floors are continuously assigned to the elevator cars, whether or not they have an active registered hall call associated therewith, with the assignments being made according to predetermined averages which uniformly spread the actual and prospective work loads among the elevator cars. The hall calls are timed. A timed-out call, i.e., a call registered for a predetermined period of time, is given preferential treatment, without significantly disturbing service to other registered hall calls, by assigning the floor and service direction associated with the timed-out call to an additional car which is not already assigned to a timed-out call. The additional car is selected on the basis of its having the lightest work load schedule of all of the elevator cars conditioned to serve the timed-out call.
Abstract:
A demand estimation apparatus for controlling machines according to revised estimated demand values wherein estimated demand values are calculated by dividing demand cycles which are fluctuating similarly cyclically into a plurality of sections (time zones), an adjusting section is interposed between each two sections adjoining each other, and an estimate of the fluctuation of the demand is determined based on measurements of both the demand value of the adjusting section and the demand values of the two adjoining sections, and wherein the estimated demand value of each adjusting section is compared with the estimated demand values of the respective two adjoining sections and, based on the comparison, each adjusting section is moved as a whole toward one of the adjoining sections by a predetermined time width when the estimated demand value of the adjusting section more closely approximates the estimated demand value of that one of the adjoining section than the other adjoining section, thus modifying the time widths of both adjoining sections, and the estimated demand value is revised to account to for fluctuations in the demand within the shifted adjusting section and modified adjoining sections.
Abstract:
An elevator system for a building having a plurality of floors, including supervisory system control for controlling a plurality of elevator cars to answer calls for elevator service from the plurality of floors. The system control assigns unassigned service directions from the plurality of floors to each of the elevator cars, until meeting a predetermined dynamic limiting average. The assignments are made to one car at a time, proceeding to the next when a predetermined limiting average is met. The order in which the cars are selected for assignment is a dynamic order, responsive to the relative work loads of the cars. The assignments are made to each car, starting in a predetermined direction from each car's position, and are terminated a predetermined travel distance from the car, if not terminated sooner due to a limiting dynamic average. A predetermined minimum limiting dynamic average may be set, to control the rate at which idle cars become busy cars when traffic increases.