Best network analysis tool ArcGIS 10.4 for out of sequence road database

[cartographercartographer]

OK I have a road database layer where all the streets have been sequenced in a non-geospatial order by segments.

These segments may divide by owner (Town, State, etc...), and/or crossing from and to street (in their own separate fields).    And then there is a road name field, and a field with an attached road segment number.  Unfortunately the road segments have NOT been arranged in geospatial sequence for the road.   So you might have a road go

road_1, road_4, road_3, road_6, road_5 if you had attached each segment together in their actual sequence they are attached together by the same road name.    I would like to use the network toolbox to get these segments all in sequence from beginning to end based on their geospatial proximity to the next segment.   What's the quickest way to do this?  The original file is a shape file with an attached DBase file.

Note: I just have a basic license.  Do I need to add the Network Analyst license, or can I still use a tool that doesn't require it from the Network Analyst Tools toolbox?

[cartographercartographer]

Because when considering two route segments that are adjacent to each other, they simplify budget planning if they are grouped together by location and common attribute, i.e. why would you want to treat two ends of a very long road when you could treat two consecutive segments of a road?   

[RichardFairhurst]

LR allows me to collapse two adjacent segments on the same route into one easily enough or more frequently to subdivide it further in response to new limits.  I deal with limit changes all the time.  Sequential numbers and sorts are far too fragile to waste my time managing them, since every new subdivision messes them up.  New subdivisions of segments has no impact on my inventories or event data.  Extending or realigning routes may, but I have better ways to readjust the event measures when routes realign or extend.  I can also detect portions of existing segments easily with LR and have no limit to the number of positions along my lines that I can reference or use as a limit, since the original lines never impose a limitation on where I can define my event segments.  I have built tools that let me define any point or segment event relative to any cross street intersection and offset distance and direction I want, whether or not the underlying lines have an end point at those positions or not.

The cross streets only have to be somewhere on the route for me to reference them.  It does not matter what underlying segments of the original network they are physically in contact with.  Therefore each and every position that exists on the route can be defined relative to every valid intersection on the route, I only have to adjust the offset distances from the fixed intersection to define them.  For facilities management and collision reporting this is ideal, since there is no way to know which side of an event a person favors or what cross streets they consider useful as a reference point.  If there are 10 cross streets on a road then there are at least 10 different valid ways to reference any position along that road relative to a cross street.

[cartographercartographer]

While I understand subdivisions can mess them up, if  a set of procedures can be put in place when a new set of roads is made, it can establish additional segments at the beginning and end by latitude and longitude.   I would not try to change the sequence of pre-existing segment that are already in sequence.

[cartographercartographer]

My problem is just trying to get an initial set of segments to work right.

[RichardFairhurst]

The methods I have outlined provide a way to get the results you want, but nothing works automatically for 100% of all roads.  All analysis tools have a 10% to 20% exception rate in my experience and you manually have to correct those exceptions.  No tool will do the work for you on 100% of all roads.  The best approach to all GIS work is to use tools to deal with the 80% to 90% of common road layouts and then invest at least 10 times the amount of effort in dealing with the exceptions.  That is the best anyone can offer.  The addition of fields is the most efficient method for being able to correct the problem cases, so that they can be scripted and reliably fixed if the process is repeated.  By doing this I am now able to regenerate my entire network of routes each week automatically to incorporate routes for all new segments.  By using the field values I am able to automatically handle each exception with a separate set of tools or steps that can all be merged at the end.  As new exceptions arise I have a well established method for classifying them and correcting them.

[cartographercartographer]

And your method does not require a Network Analyst license?  Or does it?

[DanPatterson_Retired]

the network analyst tools provide a means of establishing rules for network traversal and/or spatial delineation along and around networks.  Richard's procedures can be accomplished through judicious construction of the network in the first place.  When you construct a network data set, it does little of what he has described because that is not the prime object of the network analyst

[cartographercartographer]

OK, but does it require the Network analysis to perform these tasks?   I understand the inherent problems as I have seen them in my own analysis. 

[DanPatterson_Retired]

the network analyst isn't going to fix up a poorly designed road network or fix intersections, naming conventions, and many of the things that have been mentioned.  It is going to take what it has, explode the polylines into segments, generate intersections nodes and assess the direction of line creation.  That is about it.  Things like one-way street designation require user intervention as do turn options.  If you are trying to get a network into shape, then you do it outside of NA.  You will need it, or its equivalent, to assess whether it 'behaves' as you want it to.  When you get to the point of doing something with the network, it has a comprehensive rule set, which depends on the vector geometry and its associative attributes being correct. 

[cartographercartographer]

So in other words, there is an equal chance of being worse off with Network Analyst based results as without?