A novel methodology for pipe grouping and rehabilitation interventions scheduling in water distribution networks

Figures & data

Figure 1. Example of a pipe network descripted in (a) the classic node-arc topology, and (b) segment-valve topology.

Figure 2. Illustrative example of Tabu Search moves: (a) initial solution; (b) 1^st iteration; (c) 2^nd iteration; (d) 3^rd iteration and (e) 4^th iteration.

Figure 3. Evolution of IVI and PAC: (a) status quo situation, (b) traditional strategy in which single pipes are rehabilitated that reach their expected service life and (c) novel methodology.

Figure 4. Geospatial representation of the schedule of interventions in rehabilitation units: (a) traditional strategy in which single pipes are rehabilitated that reach their expected service life and (b) novel methodology.

Figure 5. Representation of the distribution network in the segment-valve topology with grouping pipes.

Table 1. Scenarios considered in the sensitivity analysis and obtained results.

Model assumptions			Model results
Scenario	Budget (k€)	Expected service life (year)	Number of rehabilitation units	IVI (average of the 2^nd cycle)	Average annual rehabilitation rate(%)
A	150	PG:50	72	0.35	1.4%
B		PG:75	72	0.53
C		PG:100	72	0.64
D	250	PG:50	46	0.55	2.2%
E		PG:75	46	0.70
F		PG:100	46	0.77
G		AC:40;PVC/PE:40;DI:60	46	0.56

AC: asbestos cement; PVC: polyvinyl chloride; PE: polyethylene; DI: ductile iron; PG: Pipes (general)

Figure 6. Sensitivity analysis for (a) Scenario A; (b) Scenario B; (c) Scenario C; (d) Scenario D; (e) Scenario E, (f) Scenario F and (g) Scenario G (red bars − 1^st investment cycle; green bars − 2^nd investment cycle).