Page 1 of 1
"Mutually exclusive" rule and train entry bug 27/12/2017 at 22:43 #104553 | |
Danny252
1461 posts |
There appears to be a bug related to the "Mutually exclusive" rule, which prevents one of the trains in the rule from entering even when the other train in the rule has not entered. The case I've encountered involves sidings where the train phones to enter, and I don't know if it is more generally applicable. In the attached timetables for Plymouth, two trains are included, with the first being 5Z63 to shunt from Lipson Sidings to Laira MPD, entering at 09:52. In one TT, a rule is included to make 5Z63 "Mutually exclusive" with 5Z00, a train timetabled to enter an hour later. To my understanding, this should result in 5Z63 always entering, as it runs 100% of the time and is guaranteed to enter first unless delays are turned on (this is in contrast to "Are alternatives", which would randomly choose which train is to enter). The actual effect in the timetable with the rule is that 5Z63 phones to request permission to enter, but simply does not appear after permission is given. 5Z00 does run at all (no phone call, no entry). Without the rule, 5Z63 enters as expected. Edit: The trains behave as expected when "Are alternatives" is used instead, with 5Z63 entering after the phone call (if it is chosen by the rule). Post has attachments. Log in to view them. Last edited: 27/12/2017 at 22:49 by Danny252 Reason: None given Log in to reply |
"Mutually exclusive" rule and train entry bug 28/12/2017 at 01:00 #104554 | |
GeoffM
6376 posts |
Mutually exclusive here does not mean exactly one must enter (jointly exhaustive). Here either A can enter, B can enter, or neither can enter - but not both (not jointly exhaustive). Phoning to enter and then not entering is the subject of another issue. SimSig Boss Last edited: 28/12/2017 at 01:02 by GeoffM Reason: None given Log in to reply |
"Mutually exclusive" rule and train entry bug 28/12/2017 at 01:12 #104555 | |
postal
5265 posts |
More discussion here although it dates back to 2011 so things may have changed since.
“In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Log in to reply |
"Mutually exclusive" rule and train entry bug 29/12/2017 at 11:01 #104579 | |
Danny252
1461 posts |
I'm a little unsure as to what you're saying Geoff - are you saying the phoning to enter and not appearing is not at all linked to the MutEx rule? This seems at odd with the examples I've given, where the failure of either train to appear (despite both having 100% probability) seems to be directly linked to the presence of the MutEx rule.
Log in to reply |
"Mutually exclusive" rule and train entry bug 30/12/2017 at 04:29 #104597 | |
GeoffM
6376 posts |
Danny252 in post 104579 said:I'm a little unsure as to what you're saying Geoff - are you saying the phoning to enter and not appearing is not at all linked to the MutEx rule?Correct - the phoning and not entering part happens regardless of rules. Or at least so I've been told. Now I'm wondering. This sort of thing happens, as in "X does Y" and I can't reproduce it, only to later find out that the previously unmentioned Z was a major factor too. I know that Z is not always obvious though, so I'm not complaining. SimSig Boss Log in to reply |
"Mutually exclusive" rule and train entry bug 30/12/2017 at 10:35 #104598 | |
Peter Bennet
5402 posts |
I think Mantis 13270 is relevant to this issue of phone cals ant not entering trains. On the specifics of the provided Laira add-on timetable. There are two paths 5Z01 and 5Z02 with 12 and 18 timetables respectively. All trains are set with less than 100% running so it it possible that no train will run. The MutEx rules are set to ensure that if a specific train runs then certain others will not to avoid clashes. For example if UID W1 runs then W2 and W3 will not and if W5 runs then W6 should not. I have noticed however that in my copy of this timetable the "washer" speed class is not ticked for any of the 5Z01s which is wrong. Having said all that I believe that there is now a way to automatically create random shunts but I've not explored that. Peter I identify as half man half biscuit - crumbs! Last edited: 30/12/2017 at 11:28 by Peter Bennet Reason: Wrong presumption Log in to reply |